[codex] Split backend control-plane QoL slice (#4700)
## Thinking Path
> - Paperclip is the control plane for autonomous AI companies, so
backend task ownership, recovery, review visibility, and company-scoped
limits need to stay enforceable without UI-only coupling.
> - Closed PR #4692 bundled those backend changes with UI workflow,
docs, skills, workflow, and lockfile churn.
> - PAP-2694 asks for a clean backend/control-plane slice from that
closed branch.
> - This branch starts from current `master` and mines only the `cli`,
`packages/db`, `packages/shared`, and `server` contracts/tests needed
for the backend behavior.
> - It explicitly excludes UI workflow/performance work,
`.github/workflows/pr.yml`, `pnpm-lock.yaml`, docs, skills,
package-script, adapter UI build-config, and perf fixture script
changes; the only UI files are fixture/test updates required by the
tightened shared `Company` contract.
> - The benefit is a smaller reviewable PR that preserves the
control-plane fixes while staying under Greptile s 100-file review
limit.
## What Changed
- Added company-scoped attachment-size limits through DB
schema/migrations, shared company portability contracts, CLI
import/export coverage, and server attachment upload enforcement.
- Added productivity review service/API behavior for no-comment streak,
long-active, and high-churn review issues, including request-depth
clamping and issue summary exposure.
- Hardened issue ownership and recovery/control-plane paths: peer-agent
mutation denial, issue tree pause/resume behavior, stranded recovery
origins, and related activity/test coverage.
- Preserved related backend contract updates for routine timestamp
variables and managed agent instruction bundles because they live in
shared/server contracts from the source branch.
- Addressed Greptile feedback by making `Company.attachmentMaxBytes`
non-optional, simplifying review request-depth clamping, fixing the
migration final newline, and enforcing the process-level attachment cap
as the final ceiling for uploads.
- Added minimal company fixtures needed for repo-wide typecheck/build
and kept the PR to 66 changed files with forbidden/non-slice paths
excluded.
## Verification
- `pnpm install --frozen-lockfile`
- `git diff --check origin/master..HEAD`
- `git diff --name-only origin/master..HEAD | wc -l` -> 66 files
- `git diff --name-only origin/master..HEAD -- .github/workflows/pr.yml
pnpm-lock.yaml package.json doc skills .agents scripts
packages/adapters` -> no output
- `pnpm exec vitest run --config vitest.config.ts
packages/shared/src/validators/issue.test.ts
packages/shared/src/routine-variables.test.ts
packages/shared/src/adapter-types.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
cli/src/__tests__/company.test.ts
server/src/__tests__/productivity-review-service.test.ts
server/src/__tests__/issue-tree-control-service.test.ts
server/src/__tests__/issue-tree-control-routes.test.ts
server/src/__tests__/issue-agent-mutation-ownership-routes.test.ts
server/src/__tests__/issue-attachment-routes.test.ts
server/src/__tests__/heartbeat-process-recovery.test.ts
server/src/__tests__/issues-service.test.ts` -> 12 files, 147 tests
passed
- `pnpm exec vitest run --config vitest.config.ts
cli/src/__tests__/company-delete.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
server/src/__tests__/productivity-review-service.test.ts` -> 3 files, 18
tests passed
- `pnpm exec vitest run --config vitest.config.ts
server/src/__tests__/issue-attachment-routes.test.ts` -> 1 file, 6 tests
passed
- `pnpm --filter @paperclipai/db typecheck && pnpm --filter
@paperclipai/shared typecheck && pnpm --filter @paperclipai/server
typecheck && pnpm --filter paperclipai typecheck`
- `pnpm --filter @paperclipai/server typecheck`
- `pnpm --filter @paperclipai/ui typecheck && pnpm --filter
@paperclipai/ui build`
## Risks
- Includes migrations `0073_shiny_salo.sql` and
`0074_striped_genesis.sql`; merge ordering matters if another PR adds
migrations first.
- This is intentionally backend-only apart from fixture/test updates
forced by shared type correctness; UI affordances from PR #4692 are not
present here and should land in separate UI slices.
- The worktree install emitted plugin SDK bin-link warnings for unbuilt
plugin packages, but the targeted tests and package typechecks completed
successfully.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected; check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex, GPT-5 coding agent, tool-enabled terminal/GitHub
workflow. Exact runtime context window was not exposed by the harness.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-04-28 16:46:45 -05:00
import { and , asc , desc , eq , gt , inArray , isNull , notInArray , sql } from "drizzle-orm" ;
import type { Db } from "@paperclipai/db" ;
import { clampIssueRequestDepth } from "@paperclipai/shared" ;
import {
agents ,
companies ,
costEvents ,
heartbeatRuns ,
issueComments ,
issues ,
projects ,
} from "@paperclipai/db" ;
import { logger } from "../middleware/logger.js" ;
import { logActivity } from "./activity-log.js" ;
import { budgetService } from "./budgets.js" ;
import { issueService } from "./issues.js" ;
Add recovery handoff system notices (#5289)
## Thinking Path
> - Paperclip orchestrates AI agents for zero-human companies.
> - Agent runs can end productively while the source issue still lacks a
durable final disposition.
> - That leaves the control plane unsure whether to resume, escalate, or
close the work.
> - Issue comments also need a presentation contract so system-authored
recovery notices can render as first-class thread messages without
overloading normal comments.
> - This pull request adds successful-run handoff recovery, comment
presentation metadata, and system notice rendering.
> - The benefit is stricter task liveness with clearer operator-facing
recovery state.
## What Changed
- Added successful-run handoff decisions, wake payloads, escalation
behavior, and recovery tests.
- Added issue comment presentation metadata with migration
`0078_white_darwin.sql` and shared/server/company portability support.
- Rendered recovery/system notices in issue chat with dedicated UI
components, fixtures, tests, and storybook/lab coverage.
- Included the current recovery model-profile hint patch so automatic
recovery follow-ups use the cheap profile.
## Verification
- `pnpm install --frozen-lockfile`
- `pnpm exec vitest run
server/src/services/recovery/successful-run-handoff.test.ts
ui/src/components/SystemNotice.test.tsx
ui/src/lib/system-notice-comment.test.ts
ui/src/components/IssueChatThreadSystemNotice.test.tsx`
## Risks
- Migration-bearing PR: merge this before any other branch that might
later add a migration.
- The branch touches both recovery services and issue-thread rendering,
so review should pay attention to recovery wake idempotency and comment
metadata compatibility.
## Model Used
- OpenAI GPT-5 Codex via Paperclip `codex_local` adapter, with
shell/git/GitHub CLI tool use.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-05-06 06:05:58 -05:00
import {
recoveryAssigneeAdapterOverrides ,
withRecoveryModelProfileHint ,
} from "./recovery/model-profile-hint.js" ;
[codex] Split backend control-plane QoL slice (#4700)
## Thinking Path
> - Paperclip is the control plane for autonomous AI companies, so
backend task ownership, recovery, review visibility, and company-scoped
limits need to stay enforceable without UI-only coupling.
> - Closed PR #4692 bundled those backend changes with UI workflow,
docs, skills, workflow, and lockfile churn.
> - PAP-2694 asks for a clean backend/control-plane slice from that
closed branch.
> - This branch starts from current `master` and mines only the `cli`,
`packages/db`, `packages/shared`, and `server` contracts/tests needed
for the backend behavior.
> - It explicitly excludes UI workflow/performance work,
`.github/workflows/pr.yml`, `pnpm-lock.yaml`, docs, skills,
package-script, adapter UI build-config, and perf fixture script
changes; the only UI files are fixture/test updates required by the
tightened shared `Company` contract.
> - The benefit is a smaller reviewable PR that preserves the
control-plane fixes while staying under Greptile s 100-file review
limit.
## What Changed
- Added company-scoped attachment-size limits through DB
schema/migrations, shared company portability contracts, CLI
import/export coverage, and server attachment upload enforcement.
- Added productivity review service/API behavior for no-comment streak,
long-active, and high-churn review issues, including request-depth
clamping and issue summary exposure.
- Hardened issue ownership and recovery/control-plane paths: peer-agent
mutation denial, issue tree pause/resume behavior, stranded recovery
origins, and related activity/test coverage.
- Preserved related backend contract updates for routine timestamp
variables and managed agent instruction bundles because they live in
shared/server contracts from the source branch.
- Addressed Greptile feedback by making `Company.attachmentMaxBytes`
non-optional, simplifying review request-depth clamping, fixing the
migration final newline, and enforcing the process-level attachment cap
as the final ceiling for uploads.
- Added minimal company fixtures needed for repo-wide typecheck/build
and kept the PR to 66 changed files with forbidden/non-slice paths
excluded.
## Verification
- `pnpm install --frozen-lockfile`
- `git diff --check origin/master..HEAD`
- `git diff --name-only origin/master..HEAD | wc -l` -> 66 files
- `git diff --name-only origin/master..HEAD -- .github/workflows/pr.yml
pnpm-lock.yaml package.json doc skills .agents scripts
packages/adapters` -> no output
- `pnpm exec vitest run --config vitest.config.ts
packages/shared/src/validators/issue.test.ts
packages/shared/src/routine-variables.test.ts
packages/shared/src/adapter-types.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
cli/src/__tests__/company.test.ts
server/src/__tests__/productivity-review-service.test.ts
server/src/__tests__/issue-tree-control-service.test.ts
server/src/__tests__/issue-tree-control-routes.test.ts
server/src/__tests__/issue-agent-mutation-ownership-routes.test.ts
server/src/__tests__/issue-attachment-routes.test.ts
server/src/__tests__/heartbeat-process-recovery.test.ts
server/src/__tests__/issues-service.test.ts` -> 12 files, 147 tests
passed
- `pnpm exec vitest run --config vitest.config.ts
cli/src/__tests__/company-delete.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
server/src/__tests__/productivity-review-service.test.ts` -> 3 files, 18
tests passed
- `pnpm exec vitest run --config vitest.config.ts
server/src/__tests__/issue-attachment-routes.test.ts` -> 1 file, 6 tests
passed
- `pnpm --filter @paperclipai/db typecheck && pnpm --filter
@paperclipai/shared typecheck && pnpm --filter @paperclipai/server
typecheck && pnpm --filter paperclipai typecheck`
- `pnpm --filter @paperclipai/server typecheck`
- `pnpm --filter @paperclipai/ui typecheck && pnpm --filter
@paperclipai/ui build`
## Risks
- Includes migrations `0073_shiny_salo.sql` and
`0074_striped_genesis.sql`; merge ordering matters if another PR adds
migrations first.
- This is intentionally backend-only apart from fixture/test updates
forced by shared type correctness; UI affordances from PR #4692 are not
present here and should land in separate UI slices.
- The worktree install emitted plugin SDK bin-link warnings for unbuilt
plugin packages, but the targeted tests and package typechecks completed
successfully.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected; check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex, GPT-5 coding agent, tool-enabled terminal/GitHub
workflow. Exact runtime context window was not exposed by the harness.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-04-28 16:46:45 -05:00
import { RECOVERY_ORIGIN_KINDS } from "./recovery/origins.js" ;
export const PRODUCTIVITY_REVIEW_ORIGIN_KIND = RECOVERY_ORIGIN_KINDS . issueProductivityReview ;
export const DEFAULT_PRODUCTIVITY_REVIEW_NO_COMMENT_STREAK_RUNS = 10 ;
export const DEFAULT_PRODUCTIVITY_REVIEW_LONG_ACTIVE_HOURS = 6 ;
export const DEFAULT_PRODUCTIVITY_REVIEW_HIGH_CHURN_HOURLY = 10 ;
export const DEFAULT_PRODUCTIVITY_REVIEW_HIGH_CHURN_SIX_HOURS = 30 ;
export const DEFAULT_PRODUCTIVITY_REVIEW_RESOLVED_SNOOZE_MS = 6 * 60 * 60 * 1000 ;
[codex] Bound productivity review recovery loops (#4948)
## Thinking Path
> - Paperclip orchestrates AI agents for zero-human companies.
> - The heartbeat/productivity review subsystem detects when assigned
work is likely stuck or churning.
> - Productivity reviews are useful, but repeated reconciliation can
create noisy refresh comments or repeated review issues around the same
source issue.
> - That makes manager follow-up harder because the signal can get
buried under duplicate review activity.
> - This pull request bounds productivity review refreshes and creation
loops while preserving the existing escalation path.
> - The benefit is a quieter recovery loop that still surfaces stuck or
high-churn work for manager attention.
## What Changed
- Added refresh throttling for open productivity review issues,
including a one-hour default interval and a maximum of three refresh
comments per open review.
- Added a rolling 24-hour creation cap so completed/closed reviews
cannot immediately recreate review issues indefinitely for the same
source issue.
- Excluded cancelled productivity reviews from the creation cap so
manager cancellations do not silently suppress future legitimate
reviews.
- Preserved productivity review timestamps in deterministic test paths
and added targeted coverage for immediate refresh suppression, refresh
caps, creation caps, and cancelled-review exclusion.
## Verification
- `pnpm run preflight:workspace-links && pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- `pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- Greptile Review: 5/5 on commit
`bcf25832d0ffae25890b2ee7eed112d1c2d114fe` with review threads resolved.
- GitHub PR checks passed on the latest head: `policy`, `verify`, `e2e`,
`Greptile Review`, and `security/snyk (cryppadotta)`.
- Verified the branch is rebased onto `public-gh/master` with no
conflicts.
- Verified the diff does not include `pnpm-lock.yaml`, database schema
changes, or migrations.
## Risks
- Low-to-medium risk: this changes automation cadence for productivity
reviews. A truly stuck issue may receive fewer repeated refresh
comments, but the original review issue remains open and assigned for
manager action.
- No migration risk: this is server logic and tests only.
> Checked [`ROADMAP.md`](ROADMAP.md) for overlapping planned core work;
this is a targeted recovery-loop fix and does not add a new roadmap
feature.
## Model Used
- OpenAI Codex coding agent, GPT-5 model family, tool-using software
engineering mode. Exact context window is not exposed in this runtime.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots (not applicable; server-only change)
- [x] I have updated relevant documentation to reflect my changes (not
applicable; no user-facing docs or commands changed)
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-05-01 08:32:04 -05:00
export const DEFAULT_PRODUCTIVITY_REVIEW_REFRESH_INTERVAL_MS = 60 * 60 * 1000 ;
export const DEFAULT_PRODUCTIVITY_REVIEW_MAX_REFRESH_COMMENTS = 3 ;
export const DEFAULT_PRODUCTIVITY_REVIEW_CREATION_WINDOW_MS = 24 * 60 * 60 * 1000 ;
export const DEFAULT_PRODUCTIVITY_REVIEW_MAX_CREATIONS_PER_WINDOW = 3 ;
[codex] Split backend control-plane QoL slice (#4700)
## Thinking Path
> - Paperclip is the control plane for autonomous AI companies, so
backend task ownership, recovery, review visibility, and company-scoped
limits need to stay enforceable without UI-only coupling.
> - Closed PR #4692 bundled those backend changes with UI workflow,
docs, skills, workflow, and lockfile churn.
> - PAP-2694 asks for a clean backend/control-plane slice from that
closed branch.
> - This branch starts from current `master` and mines only the `cli`,
`packages/db`, `packages/shared`, and `server` contracts/tests needed
for the backend behavior.
> - It explicitly excludes UI workflow/performance work,
`.github/workflows/pr.yml`, `pnpm-lock.yaml`, docs, skills,
package-script, adapter UI build-config, and perf fixture script
changes; the only UI files are fixture/test updates required by the
tightened shared `Company` contract.
> - The benefit is a smaller reviewable PR that preserves the
control-plane fixes while staying under Greptile s 100-file review
limit.
## What Changed
- Added company-scoped attachment-size limits through DB
schema/migrations, shared company portability contracts, CLI
import/export coverage, and server attachment upload enforcement.
- Added productivity review service/API behavior for no-comment streak,
long-active, and high-churn review issues, including request-depth
clamping and issue summary exposure.
- Hardened issue ownership and recovery/control-plane paths: peer-agent
mutation denial, issue tree pause/resume behavior, stranded recovery
origins, and related activity/test coverage.
- Preserved related backend contract updates for routine timestamp
variables and managed agent instruction bundles because they live in
shared/server contracts from the source branch.
- Addressed Greptile feedback by making `Company.attachmentMaxBytes`
non-optional, simplifying review request-depth clamping, fixing the
migration final newline, and enforcing the process-level attachment cap
as the final ceiling for uploads.
- Added minimal company fixtures needed for repo-wide typecheck/build
and kept the PR to 66 changed files with forbidden/non-slice paths
excluded.
## Verification
- `pnpm install --frozen-lockfile`
- `git diff --check origin/master..HEAD`
- `git diff --name-only origin/master..HEAD | wc -l` -> 66 files
- `git diff --name-only origin/master..HEAD -- .github/workflows/pr.yml
pnpm-lock.yaml package.json doc skills .agents scripts
packages/adapters` -> no output
- `pnpm exec vitest run --config vitest.config.ts
packages/shared/src/validators/issue.test.ts
packages/shared/src/routine-variables.test.ts
packages/shared/src/adapter-types.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
cli/src/__tests__/company.test.ts
server/src/__tests__/productivity-review-service.test.ts
server/src/__tests__/issue-tree-control-service.test.ts
server/src/__tests__/issue-tree-control-routes.test.ts
server/src/__tests__/issue-agent-mutation-ownership-routes.test.ts
server/src/__tests__/issue-attachment-routes.test.ts
server/src/__tests__/heartbeat-process-recovery.test.ts
server/src/__tests__/issues-service.test.ts` -> 12 files, 147 tests
passed
- `pnpm exec vitest run --config vitest.config.ts
cli/src/__tests__/company-delete.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
server/src/__tests__/productivity-review-service.test.ts` -> 3 files, 18
tests passed
- `pnpm exec vitest run --config vitest.config.ts
server/src/__tests__/issue-attachment-routes.test.ts` -> 1 file, 6 tests
passed
- `pnpm --filter @paperclipai/db typecheck && pnpm --filter
@paperclipai/shared typecheck && pnpm --filter @paperclipai/server
typecheck && pnpm --filter paperclipai typecheck`
- `pnpm --filter @paperclipai/server typecheck`
- `pnpm --filter @paperclipai/ui typecheck && pnpm --filter
@paperclipai/ui build`
## Risks
- Includes migrations `0073_shiny_salo.sql` and
`0074_striped_genesis.sql`; merge ordering matters if another PR adds
migrations first.
- This is intentionally backend-only apart from fixture/test updates
forced by shared type correctness; UI affordances from PR #4692 are not
present here and should land in separate UI slices.
- The worktree install emitted plugin SDK bin-link warnings for unbuilt
plugin packages, but the targeted tests and package typechecks completed
successfully.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected; check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex, GPT-5 coding agent, tool-enabled terminal/GitHub
workflow. Exact runtime context window was not exposed by the harness.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-04-28 16:46:45 -05:00
const TERMINAL_RUN_STATUSES = [ "succeeded" , "failed" , "cancelled" , "timed_out" ] as const ;
const ACTIVE_RUN_STATUSES = [ "queued" , "running" , "scheduled_retry" ] as const ;
const MAX_CANDIDATE_ISSUES = 250 ;
const MAX_RUNS_FOR_STREAK = 100 ;
const MAX_PARENT_WALK_DEPTH = 25 ;
[codex] Bound productivity review recovery loops (#4948)
## Thinking Path
> - Paperclip orchestrates AI agents for zero-human companies.
> - The heartbeat/productivity review subsystem detects when assigned
work is likely stuck or churning.
> - Productivity reviews are useful, but repeated reconciliation can
create noisy refresh comments or repeated review issues around the same
source issue.
> - That makes manager follow-up harder because the signal can get
buried under duplicate review activity.
> - This pull request bounds productivity review refreshes and creation
loops while preserving the existing escalation path.
> - The benefit is a quieter recovery loop that still surfaces stuck or
high-churn work for manager attention.
## What Changed
- Added refresh throttling for open productivity review issues,
including a one-hour default interval and a maximum of three refresh
comments per open review.
- Added a rolling 24-hour creation cap so completed/closed reviews
cannot immediately recreate review issues indefinitely for the same
source issue.
- Excluded cancelled productivity reviews from the creation cap so
manager cancellations do not silently suppress future legitimate
reviews.
- Preserved productivity review timestamps in deterministic test paths
and added targeted coverage for immediate refresh suppression, refresh
caps, creation caps, and cancelled-review exclusion.
## Verification
- `pnpm run preflight:workspace-links && pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- `pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- Greptile Review: 5/5 on commit
`bcf25832d0ffae25890b2ee7eed112d1c2d114fe` with review threads resolved.
- GitHub PR checks passed on the latest head: `policy`, `verify`, `e2e`,
`Greptile Review`, and `security/snyk (cryppadotta)`.
- Verified the branch is rebased onto `public-gh/master` with no
conflicts.
- Verified the diff does not include `pnpm-lock.yaml`, database schema
changes, or migrations.
## Risks
- Low-to-medium risk: this changes automation cadence for productivity
reviews. A truly stuck issue may receive fewer repeated refresh
comments, but the original review issue remains open and assigned for
manager action.
- No migration risk: this is server logic and tests only.
> Checked [`ROADMAP.md`](ROADMAP.md) for overlapping planned core work;
this is a targeted recovery-loop fix and does not add a new roadmap
feature.
## Model Used
- OpenAI Codex coding agent, GPT-5 model family, tool-using software
engineering mode. Exact context window is not exposed in this runtime.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots (not applicable; server-only change)
- [x] I have updated relevant documentation to reflect my changes (not
applicable; no user-facing docs or commands changed)
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-05-01 08:32:04 -05:00
export const PRODUCTIVITY_REVIEW_REFRESH_COMMENT_PREFIX = "Productivity review evidence refreshed." ;
[codex] Split backend control-plane QoL slice (#4700)
## Thinking Path
> - Paperclip is the control plane for autonomous AI companies, so
backend task ownership, recovery, review visibility, and company-scoped
limits need to stay enforceable without UI-only coupling.
> - Closed PR #4692 bundled those backend changes with UI workflow,
docs, skills, workflow, and lockfile churn.
> - PAP-2694 asks for a clean backend/control-plane slice from that
closed branch.
> - This branch starts from current `master` and mines only the `cli`,
`packages/db`, `packages/shared`, and `server` contracts/tests needed
for the backend behavior.
> - It explicitly excludes UI workflow/performance work,
`.github/workflows/pr.yml`, `pnpm-lock.yaml`, docs, skills,
package-script, adapter UI build-config, and perf fixture script
changes; the only UI files are fixture/test updates required by the
tightened shared `Company` contract.
> - The benefit is a smaller reviewable PR that preserves the
control-plane fixes while staying under Greptile s 100-file review
limit.
## What Changed
- Added company-scoped attachment-size limits through DB
schema/migrations, shared company portability contracts, CLI
import/export coverage, and server attachment upload enforcement.
- Added productivity review service/API behavior for no-comment streak,
long-active, and high-churn review issues, including request-depth
clamping and issue summary exposure.
- Hardened issue ownership and recovery/control-plane paths: peer-agent
mutation denial, issue tree pause/resume behavior, stranded recovery
origins, and related activity/test coverage.
- Preserved related backend contract updates for routine timestamp
variables and managed agent instruction bundles because they live in
shared/server contracts from the source branch.
- Addressed Greptile feedback by making `Company.attachmentMaxBytes`
non-optional, simplifying review request-depth clamping, fixing the
migration final newline, and enforcing the process-level attachment cap
as the final ceiling for uploads.
- Added minimal company fixtures needed for repo-wide typecheck/build
and kept the PR to 66 changed files with forbidden/non-slice paths
excluded.
## Verification
- `pnpm install --frozen-lockfile`
- `git diff --check origin/master..HEAD`
- `git diff --name-only origin/master..HEAD | wc -l` -> 66 files
- `git diff --name-only origin/master..HEAD -- .github/workflows/pr.yml
pnpm-lock.yaml package.json doc skills .agents scripts
packages/adapters` -> no output
- `pnpm exec vitest run --config vitest.config.ts
packages/shared/src/validators/issue.test.ts
packages/shared/src/routine-variables.test.ts
packages/shared/src/adapter-types.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
cli/src/__tests__/company.test.ts
server/src/__tests__/productivity-review-service.test.ts
server/src/__tests__/issue-tree-control-service.test.ts
server/src/__tests__/issue-tree-control-routes.test.ts
server/src/__tests__/issue-agent-mutation-ownership-routes.test.ts
server/src/__tests__/issue-attachment-routes.test.ts
server/src/__tests__/heartbeat-process-recovery.test.ts
server/src/__tests__/issues-service.test.ts` -> 12 files, 147 tests
passed
- `pnpm exec vitest run --config vitest.config.ts
cli/src/__tests__/company-delete.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
server/src/__tests__/productivity-review-service.test.ts` -> 3 files, 18
tests passed
- `pnpm exec vitest run --config vitest.config.ts
server/src/__tests__/issue-attachment-routes.test.ts` -> 1 file, 6 tests
passed
- `pnpm --filter @paperclipai/db typecheck && pnpm --filter
@paperclipai/shared typecheck && pnpm --filter @paperclipai/server
typecheck && pnpm --filter paperclipai typecheck`
- `pnpm --filter @paperclipai/server typecheck`
- `pnpm --filter @paperclipai/ui typecheck && pnpm --filter
@paperclipai/ui build`
## Risks
- Includes migrations `0073_shiny_salo.sql` and
`0074_striped_genesis.sql`; merge ordering matters if another PR adds
migrations first.
- This is intentionally backend-only apart from fixture/test updates
forced by shared type correctness; UI affordances from PR #4692 are not
present here and should land in separate UI slices.
- The worktree install emitted plugin SDK bin-link warnings for unbuilt
plugin packages, but the targeted tests and package typechecks completed
successfully.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected; check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex, GPT-5 coding agent, tool-enabled terminal/GitHub
workflow. Exact runtime context window was not exposed by the harness.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-04-28 16:46:45 -05:00
type IssueRow = typeof issues . $inferSelect ;
type AgentRow = typeof agents . $inferSelect ;
type HeartbeatRunRow = typeof heartbeatRuns . $inferSelect ;
type ProductivityReviewTrigger = "no_comment_streak" | "long_active_duration" | "high_churn" ;
type ProductivityReviewThresholds = {
noCommentStreakRuns : number ;
longActiveMs : number ;
highChurnHourly : number ;
highChurnSixHours : number ;
resolvedSnoozeMs : number ;
[codex] Bound productivity review recovery loops (#4948)
## Thinking Path
> - Paperclip orchestrates AI agents for zero-human companies.
> - The heartbeat/productivity review subsystem detects when assigned
work is likely stuck or churning.
> - Productivity reviews are useful, but repeated reconciliation can
create noisy refresh comments or repeated review issues around the same
source issue.
> - That makes manager follow-up harder because the signal can get
buried under duplicate review activity.
> - This pull request bounds productivity review refreshes and creation
loops while preserving the existing escalation path.
> - The benefit is a quieter recovery loop that still surfaces stuck or
high-churn work for manager attention.
## What Changed
- Added refresh throttling for open productivity review issues,
including a one-hour default interval and a maximum of three refresh
comments per open review.
- Added a rolling 24-hour creation cap so completed/closed reviews
cannot immediately recreate review issues indefinitely for the same
source issue.
- Excluded cancelled productivity reviews from the creation cap so
manager cancellations do not silently suppress future legitimate
reviews.
- Preserved productivity review timestamps in deterministic test paths
and added targeted coverage for immediate refresh suppression, refresh
caps, creation caps, and cancelled-review exclusion.
## Verification
- `pnpm run preflight:workspace-links && pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- `pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- Greptile Review: 5/5 on commit
`bcf25832d0ffae25890b2ee7eed112d1c2d114fe` with review threads resolved.
- GitHub PR checks passed on the latest head: `policy`, `verify`, `e2e`,
`Greptile Review`, and `security/snyk (cryppadotta)`.
- Verified the branch is rebased onto `public-gh/master` with no
conflicts.
- Verified the diff does not include `pnpm-lock.yaml`, database schema
changes, or migrations.
## Risks
- Low-to-medium risk: this changes automation cadence for productivity
reviews. A truly stuck issue may receive fewer repeated refresh
comments, but the original review issue remains open and assigned for
manager action.
- No migration risk: this is server logic and tests only.
> Checked [`ROADMAP.md`](ROADMAP.md) for overlapping planned core work;
this is a targeted recovery-loop fix and does not add a new roadmap
feature.
## Model Used
- OpenAI Codex coding agent, GPT-5 model family, tool-using software
engineering mode. Exact context window is not exposed in this runtime.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots (not applicable; server-only change)
- [x] I have updated relevant documentation to reflect my changes (not
applicable; no user-facing docs or commands changed)
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-05-01 08:32:04 -05:00
refreshIntervalMs : number ;
maxRefreshComments : number ;
creationWindowMs : number ;
maxCreationsPerWindow : number ;
[codex] Split backend control-plane QoL slice (#4700)
## Thinking Path
> - Paperclip is the control plane for autonomous AI companies, so
backend task ownership, recovery, review visibility, and company-scoped
limits need to stay enforceable without UI-only coupling.
> - Closed PR #4692 bundled those backend changes with UI workflow,
docs, skills, workflow, and lockfile churn.
> - PAP-2694 asks for a clean backend/control-plane slice from that
closed branch.
> - This branch starts from current `master` and mines only the `cli`,
`packages/db`, `packages/shared`, and `server` contracts/tests needed
for the backend behavior.
> - It explicitly excludes UI workflow/performance work,
`.github/workflows/pr.yml`, `pnpm-lock.yaml`, docs, skills,
package-script, adapter UI build-config, and perf fixture script
changes; the only UI files are fixture/test updates required by the
tightened shared `Company` contract.
> - The benefit is a smaller reviewable PR that preserves the
control-plane fixes while staying under Greptile s 100-file review
limit.
## What Changed
- Added company-scoped attachment-size limits through DB
schema/migrations, shared company portability contracts, CLI
import/export coverage, and server attachment upload enforcement.
- Added productivity review service/API behavior for no-comment streak,
long-active, and high-churn review issues, including request-depth
clamping and issue summary exposure.
- Hardened issue ownership and recovery/control-plane paths: peer-agent
mutation denial, issue tree pause/resume behavior, stranded recovery
origins, and related activity/test coverage.
- Preserved related backend contract updates for routine timestamp
variables and managed agent instruction bundles because they live in
shared/server contracts from the source branch.
- Addressed Greptile feedback by making `Company.attachmentMaxBytes`
non-optional, simplifying review request-depth clamping, fixing the
migration final newline, and enforcing the process-level attachment cap
as the final ceiling for uploads.
- Added minimal company fixtures needed for repo-wide typecheck/build
and kept the PR to 66 changed files with forbidden/non-slice paths
excluded.
## Verification
- `pnpm install --frozen-lockfile`
- `git diff --check origin/master..HEAD`
- `git diff --name-only origin/master..HEAD | wc -l` -> 66 files
- `git diff --name-only origin/master..HEAD -- .github/workflows/pr.yml
pnpm-lock.yaml package.json doc skills .agents scripts
packages/adapters` -> no output
- `pnpm exec vitest run --config vitest.config.ts
packages/shared/src/validators/issue.test.ts
packages/shared/src/routine-variables.test.ts
packages/shared/src/adapter-types.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
cli/src/__tests__/company.test.ts
server/src/__tests__/productivity-review-service.test.ts
server/src/__tests__/issue-tree-control-service.test.ts
server/src/__tests__/issue-tree-control-routes.test.ts
server/src/__tests__/issue-agent-mutation-ownership-routes.test.ts
server/src/__tests__/issue-attachment-routes.test.ts
server/src/__tests__/heartbeat-process-recovery.test.ts
server/src/__tests__/issues-service.test.ts` -> 12 files, 147 tests
passed
- `pnpm exec vitest run --config vitest.config.ts
cli/src/__tests__/company-delete.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
server/src/__tests__/productivity-review-service.test.ts` -> 3 files, 18
tests passed
- `pnpm exec vitest run --config vitest.config.ts
server/src/__tests__/issue-attachment-routes.test.ts` -> 1 file, 6 tests
passed
- `pnpm --filter @paperclipai/db typecheck && pnpm --filter
@paperclipai/shared typecheck && pnpm --filter @paperclipai/server
typecheck && pnpm --filter paperclipai typecheck`
- `pnpm --filter @paperclipai/server typecheck`
- `pnpm --filter @paperclipai/ui typecheck && pnpm --filter
@paperclipai/ui build`
## Risks
- Includes migrations `0073_shiny_salo.sql` and
`0074_striped_genesis.sql`; merge ordering matters if another PR adds
migrations first.
- This is intentionally backend-only apart from fixture/test updates
forced by shared type correctness; UI affordances from PR #4692 are not
present here and should land in separate UI slices.
- The worktree install emitted plugin SDK bin-link warnings for unbuilt
plugin packages, but the targeted tests and package typechecks completed
successfully.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected; check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex, GPT-5 coding agent, tool-enabled terminal/GitHub
workflow. Exact runtime context window was not exposed by the harness.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-04-28 16:46:45 -05:00
} ;
type ProductivityReviewEvidence = {
trigger : ProductivityReviewTrigger ;
triggerReasons : string [ ] ;
sourceIssue : IssueRow ;
sourceAgent : AgentRow ;
noCommentStreak : number ;
totalRunCount : number ;
terminalRunCount : number ;
activeRunCount : number ;
runCountLastHour : number ;
runCountLastSixHours : number ;
commentCount : number ;
commentCountLastHour : number ;
commentCountLastSixHours : number ;
elapsedMs : number | null ;
latestRuns : HeartbeatRunRow [ ] ;
latestComments : Array < typeof issueComments. $ inferSelect > ;
costCents : number ;
usageSamples : Array < { runId : string ; usageJson : Record < string , unknown > | null } > ;
nextAction : string | null ;
thresholds : ProductivityReviewThresholds ;
generatedAt : Date ;
} ;
type EnqueueWakeup = (
agentId : string ,
opts ? : {
source ? : "timer" | "assignment" | "on_demand" | "automation" ;
triggerDetail ? : "manual" | "ping" | "callback" | "system" ;
reason? : string | null ;
payload? : Record < string , unknown > | null ;
requestedByActorType ? : "user" | "agent" | "system" ;
requestedByActorId? : string | null ;
contextSnapshot? : Record < string , unknown > ;
} ,
) = > Promise < unknown | null > ;
function productivityReviewFingerprint ( sourceIssueId : string ) {
return ` productivity-review: ${ sourceIssueId } ` ;
}
function issueRunScopeSql ( issueId : string ) {
return sql ` (
$ { heartbeatRuns . contextSnapshot } - >> 'issueId' = $ { issueId }
or $ { heartbeatRuns . contextSnapshot } - >> 'taskId' = $ { issueId }
or $ { heartbeatRuns . contextSnapshot } - >> 'taskKey' = $ { issueId }
) ` ;
}
function msToHuman ( ms : number | null ) {
if ( ms === null ) return "unknown" ;
const minutes = Math . floor ( ms / 60 _000 ) ;
if ( minutes < 60 ) return ` ${ minutes } m ` ;
const hours = Math . floor ( minutes / 60 ) ;
const days = Math . floor ( hours / 24 ) ;
if ( days > 0 ) return ` ${ days } d ${ hours % 24 } h ` ;
return ` ${ hours } h ${ minutes % 60 } m ` ;
}
function issueUiLink ( issue : { identifier : string | null ; id : string } , prefix : string ) {
const label = issue . identifier ? ? issue . id ;
return ` [ ${ label } ](/ ${ prefix } /issues/ ${ label } ) ` ;
}
function runUiLink ( run : { id : string ; agentId : string } , prefix : string ) {
return ` [ ${ run . id } ](/ ${ prefix } /agents/ ${ run . agentId } /runs/ ${ run . id } ) ` ;
}
function truncateInline ( value : string | null | undefined , max = 260 ) {
if ( ! value ) return "" ;
const compact = value . replace ( /\s+/g , " " ) . trim ( ) ;
return compact . length <= max ? compact : ` ${ compact . slice ( 0 , max - 3 ) } ... ` ;
}
function readPositiveInteger ( value : number , fallback : number ) {
return Number . isFinite ( value ) && value > 0 ? Math . floor ( value ) : fallback ;
}
[codex] Bound productivity review recovery loops (#4948)
## Thinking Path
> - Paperclip orchestrates AI agents for zero-human companies.
> - The heartbeat/productivity review subsystem detects when assigned
work is likely stuck or churning.
> - Productivity reviews are useful, but repeated reconciliation can
create noisy refresh comments or repeated review issues around the same
source issue.
> - That makes manager follow-up harder because the signal can get
buried under duplicate review activity.
> - This pull request bounds productivity review refreshes and creation
loops while preserving the existing escalation path.
> - The benefit is a quieter recovery loop that still surfaces stuck or
high-churn work for manager attention.
## What Changed
- Added refresh throttling for open productivity review issues,
including a one-hour default interval and a maximum of three refresh
comments per open review.
- Added a rolling 24-hour creation cap so completed/closed reviews
cannot immediately recreate review issues indefinitely for the same
source issue.
- Excluded cancelled productivity reviews from the creation cap so
manager cancellations do not silently suppress future legitimate
reviews.
- Preserved productivity review timestamps in deterministic test paths
and added targeted coverage for immediate refresh suppression, refresh
caps, creation caps, and cancelled-review exclusion.
## Verification
- `pnpm run preflight:workspace-links && pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- `pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- Greptile Review: 5/5 on commit
`bcf25832d0ffae25890b2ee7eed112d1c2d114fe` with review threads resolved.
- GitHub PR checks passed on the latest head: `policy`, `verify`, `e2e`,
`Greptile Review`, and `security/snyk (cryppadotta)`.
- Verified the branch is rebased onto `public-gh/master` with no
conflicts.
- Verified the diff does not include `pnpm-lock.yaml`, database schema
changes, or migrations.
## Risks
- Low-to-medium risk: this changes automation cadence for productivity
reviews. A truly stuck issue may receive fewer repeated refresh
comments, but the original review issue remains open and assigned for
manager action.
- No migration risk: this is server logic and tests only.
> Checked [`ROADMAP.md`](ROADMAP.md) for overlapping planned core work;
this is a targeted recovery-loop fix and does not add a new roadmap
feature.
## Model Used
- OpenAI Codex coding agent, GPT-5 model family, tool-using software
engineering mode. Exact context window is not exposed in this runtime.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots (not applicable; server-only change)
- [x] I have updated relevant documentation to reflect my changes (not
applicable; no user-facing docs or commands changed)
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-05-01 08:32:04 -05:00
function coerceDate ( value : Date | string | null | undefined ) {
if ( ! value ) return null ;
return value instanceof Date ? value : new Date ( value ) ;
}
[codex] Split backend control-plane QoL slice (#4700)
## Thinking Path
> - Paperclip is the control plane for autonomous AI companies, so
backend task ownership, recovery, review visibility, and company-scoped
limits need to stay enforceable without UI-only coupling.
> - Closed PR #4692 bundled those backend changes with UI workflow,
docs, skills, workflow, and lockfile churn.
> - PAP-2694 asks for a clean backend/control-plane slice from that
closed branch.
> - This branch starts from current `master` and mines only the `cli`,
`packages/db`, `packages/shared`, and `server` contracts/tests needed
for the backend behavior.
> - It explicitly excludes UI workflow/performance work,
`.github/workflows/pr.yml`, `pnpm-lock.yaml`, docs, skills,
package-script, adapter UI build-config, and perf fixture script
changes; the only UI files are fixture/test updates required by the
tightened shared `Company` contract.
> - The benefit is a smaller reviewable PR that preserves the
control-plane fixes while staying under Greptile s 100-file review
limit.
## What Changed
- Added company-scoped attachment-size limits through DB
schema/migrations, shared company portability contracts, CLI
import/export coverage, and server attachment upload enforcement.
- Added productivity review service/API behavior for no-comment streak,
long-active, and high-churn review issues, including request-depth
clamping and issue summary exposure.
- Hardened issue ownership and recovery/control-plane paths: peer-agent
mutation denial, issue tree pause/resume behavior, stranded recovery
origins, and related activity/test coverage.
- Preserved related backend contract updates for routine timestamp
variables and managed agent instruction bundles because they live in
shared/server contracts from the source branch.
- Addressed Greptile feedback by making `Company.attachmentMaxBytes`
non-optional, simplifying review request-depth clamping, fixing the
migration final newline, and enforcing the process-level attachment cap
as the final ceiling for uploads.
- Added minimal company fixtures needed for repo-wide typecheck/build
and kept the PR to 66 changed files with forbidden/non-slice paths
excluded.
## Verification
- `pnpm install --frozen-lockfile`
- `git diff --check origin/master..HEAD`
- `git diff --name-only origin/master..HEAD | wc -l` -> 66 files
- `git diff --name-only origin/master..HEAD -- .github/workflows/pr.yml
pnpm-lock.yaml package.json doc skills .agents scripts
packages/adapters` -> no output
- `pnpm exec vitest run --config vitest.config.ts
packages/shared/src/validators/issue.test.ts
packages/shared/src/routine-variables.test.ts
packages/shared/src/adapter-types.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
cli/src/__tests__/company.test.ts
server/src/__tests__/productivity-review-service.test.ts
server/src/__tests__/issue-tree-control-service.test.ts
server/src/__tests__/issue-tree-control-routes.test.ts
server/src/__tests__/issue-agent-mutation-ownership-routes.test.ts
server/src/__tests__/issue-attachment-routes.test.ts
server/src/__tests__/heartbeat-process-recovery.test.ts
server/src/__tests__/issues-service.test.ts` -> 12 files, 147 tests
passed
- `pnpm exec vitest run --config vitest.config.ts
cli/src/__tests__/company-delete.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
server/src/__tests__/productivity-review-service.test.ts` -> 3 files, 18
tests passed
- `pnpm exec vitest run --config vitest.config.ts
server/src/__tests__/issue-attachment-routes.test.ts` -> 1 file, 6 tests
passed
- `pnpm --filter @paperclipai/db typecheck && pnpm --filter
@paperclipai/shared typecheck && pnpm --filter @paperclipai/server
typecheck && pnpm --filter paperclipai typecheck`
- `pnpm --filter @paperclipai/server typecheck`
- `pnpm --filter @paperclipai/ui typecheck && pnpm --filter
@paperclipai/ui build`
## Risks
- Includes migrations `0073_shiny_salo.sql` and
`0074_striped_genesis.sql`; merge ordering matters if another PR adds
migrations first.
- This is intentionally backend-only apart from fixture/test updates
forced by shared type correctness; UI affordances from PR #4692 are not
present here and should land in separate UI slices.
- The worktree install emitted plugin SDK bin-link warnings for unbuilt
plugin packages, but the targeted tests and package typechecks completed
successfully.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected; check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex, GPT-5 coding agent, tool-enabled terminal/GitHub
workflow. Exact runtime context window was not exposed by the harness.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-04-28 16:46:45 -05:00
function buildThresholds ( overrides? : Partial < ProductivityReviewThresholds > ) : ProductivityReviewThresholds {
return {
noCommentStreakRuns : readPositiveInteger (
overrides ? . noCommentStreakRuns ? ? DEFAULT_PRODUCTIVITY_REVIEW_NO_COMMENT_STREAK_RUNS ,
DEFAULT_PRODUCTIVITY_REVIEW_NO_COMMENT_STREAK_RUNS ,
) ,
longActiveMs : readPositiveInteger (
overrides ? . longActiveMs ? ? DEFAULT_PRODUCTIVITY_REVIEW_LONG_ACTIVE_HOURS * 60 * 60 * 1000 ,
DEFAULT_PRODUCTIVITY_REVIEW_LONG_ACTIVE_HOURS * 60 * 60 * 1000 ,
) ,
highChurnHourly : readPositiveInteger (
overrides ? . highChurnHourly ? ? DEFAULT_PRODUCTIVITY_REVIEW_HIGH_CHURN_HOURLY ,
DEFAULT_PRODUCTIVITY_REVIEW_HIGH_CHURN_HOURLY ,
) ,
highChurnSixHours : readPositiveInteger (
overrides ? . highChurnSixHours ? ? DEFAULT_PRODUCTIVITY_REVIEW_HIGH_CHURN_SIX_HOURS ,
DEFAULT_PRODUCTIVITY_REVIEW_HIGH_CHURN_SIX_HOURS ,
) ,
resolvedSnoozeMs : readPositiveInteger (
overrides ? . resolvedSnoozeMs ? ? DEFAULT_PRODUCTIVITY_REVIEW_RESOLVED_SNOOZE_MS ,
DEFAULT_PRODUCTIVITY_REVIEW_RESOLVED_SNOOZE_MS ,
) ,
[codex] Bound productivity review recovery loops (#4948)
## Thinking Path
> - Paperclip orchestrates AI agents for zero-human companies.
> - The heartbeat/productivity review subsystem detects when assigned
work is likely stuck or churning.
> - Productivity reviews are useful, but repeated reconciliation can
create noisy refresh comments or repeated review issues around the same
source issue.
> - That makes manager follow-up harder because the signal can get
buried under duplicate review activity.
> - This pull request bounds productivity review refreshes and creation
loops while preserving the existing escalation path.
> - The benefit is a quieter recovery loop that still surfaces stuck or
high-churn work for manager attention.
## What Changed
- Added refresh throttling for open productivity review issues,
including a one-hour default interval and a maximum of three refresh
comments per open review.
- Added a rolling 24-hour creation cap so completed/closed reviews
cannot immediately recreate review issues indefinitely for the same
source issue.
- Excluded cancelled productivity reviews from the creation cap so
manager cancellations do not silently suppress future legitimate
reviews.
- Preserved productivity review timestamps in deterministic test paths
and added targeted coverage for immediate refresh suppression, refresh
caps, creation caps, and cancelled-review exclusion.
## Verification
- `pnpm run preflight:workspace-links && pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- `pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- Greptile Review: 5/5 on commit
`bcf25832d0ffae25890b2ee7eed112d1c2d114fe` with review threads resolved.
- GitHub PR checks passed on the latest head: `policy`, `verify`, `e2e`,
`Greptile Review`, and `security/snyk (cryppadotta)`.
- Verified the branch is rebased onto `public-gh/master` with no
conflicts.
- Verified the diff does not include `pnpm-lock.yaml`, database schema
changes, or migrations.
## Risks
- Low-to-medium risk: this changes automation cadence for productivity
reviews. A truly stuck issue may receive fewer repeated refresh
comments, but the original review issue remains open and assigned for
manager action.
- No migration risk: this is server logic and tests only.
> Checked [`ROADMAP.md`](ROADMAP.md) for overlapping planned core work;
this is a targeted recovery-loop fix and does not add a new roadmap
feature.
## Model Used
- OpenAI Codex coding agent, GPT-5 model family, tool-using software
engineering mode. Exact context window is not exposed in this runtime.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots (not applicable; server-only change)
- [x] I have updated relevant documentation to reflect my changes (not
applicable; no user-facing docs or commands changed)
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-05-01 08:32:04 -05:00
refreshIntervalMs : readPositiveInteger (
overrides ? . refreshIntervalMs ? ? DEFAULT_PRODUCTIVITY_REVIEW_REFRESH_INTERVAL_MS ,
DEFAULT_PRODUCTIVITY_REVIEW_REFRESH_INTERVAL_MS ,
) ,
maxRefreshComments : readPositiveInteger (
overrides ? . maxRefreshComments ? ? DEFAULT_PRODUCTIVITY_REVIEW_MAX_REFRESH_COMMENTS ,
DEFAULT_PRODUCTIVITY_REVIEW_MAX_REFRESH_COMMENTS ,
) ,
creationWindowMs : readPositiveInteger (
overrides ? . creationWindowMs ? ? DEFAULT_PRODUCTIVITY_REVIEW_CREATION_WINDOW_MS ,
DEFAULT_PRODUCTIVITY_REVIEW_CREATION_WINDOW_MS ,
) ,
maxCreationsPerWindow : readPositiveInteger (
overrides ? . maxCreationsPerWindow ? ? DEFAULT_PRODUCTIVITY_REVIEW_MAX_CREATIONS_PER_WINDOW ,
DEFAULT_PRODUCTIVITY_REVIEW_MAX_CREATIONS_PER_WINDOW ,
) ,
[codex] Split backend control-plane QoL slice (#4700)
## Thinking Path
> - Paperclip is the control plane for autonomous AI companies, so
backend task ownership, recovery, review visibility, and company-scoped
limits need to stay enforceable without UI-only coupling.
> - Closed PR #4692 bundled those backend changes with UI workflow,
docs, skills, workflow, and lockfile churn.
> - PAP-2694 asks for a clean backend/control-plane slice from that
closed branch.
> - This branch starts from current `master` and mines only the `cli`,
`packages/db`, `packages/shared`, and `server` contracts/tests needed
for the backend behavior.
> - It explicitly excludes UI workflow/performance work,
`.github/workflows/pr.yml`, `pnpm-lock.yaml`, docs, skills,
package-script, adapter UI build-config, and perf fixture script
changes; the only UI files are fixture/test updates required by the
tightened shared `Company` contract.
> - The benefit is a smaller reviewable PR that preserves the
control-plane fixes while staying under Greptile s 100-file review
limit.
## What Changed
- Added company-scoped attachment-size limits through DB
schema/migrations, shared company portability contracts, CLI
import/export coverage, and server attachment upload enforcement.
- Added productivity review service/API behavior for no-comment streak,
long-active, and high-churn review issues, including request-depth
clamping and issue summary exposure.
- Hardened issue ownership and recovery/control-plane paths: peer-agent
mutation denial, issue tree pause/resume behavior, stranded recovery
origins, and related activity/test coverage.
- Preserved related backend contract updates for routine timestamp
variables and managed agent instruction bundles because they live in
shared/server contracts from the source branch.
- Addressed Greptile feedback by making `Company.attachmentMaxBytes`
non-optional, simplifying review request-depth clamping, fixing the
migration final newline, and enforcing the process-level attachment cap
as the final ceiling for uploads.
- Added minimal company fixtures needed for repo-wide typecheck/build
and kept the PR to 66 changed files with forbidden/non-slice paths
excluded.
## Verification
- `pnpm install --frozen-lockfile`
- `git diff --check origin/master..HEAD`
- `git diff --name-only origin/master..HEAD | wc -l` -> 66 files
- `git diff --name-only origin/master..HEAD -- .github/workflows/pr.yml
pnpm-lock.yaml package.json doc skills .agents scripts
packages/adapters` -> no output
- `pnpm exec vitest run --config vitest.config.ts
packages/shared/src/validators/issue.test.ts
packages/shared/src/routine-variables.test.ts
packages/shared/src/adapter-types.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
cli/src/__tests__/company.test.ts
server/src/__tests__/productivity-review-service.test.ts
server/src/__tests__/issue-tree-control-service.test.ts
server/src/__tests__/issue-tree-control-routes.test.ts
server/src/__tests__/issue-agent-mutation-ownership-routes.test.ts
server/src/__tests__/issue-attachment-routes.test.ts
server/src/__tests__/heartbeat-process-recovery.test.ts
server/src/__tests__/issues-service.test.ts` -> 12 files, 147 tests
passed
- `pnpm exec vitest run --config vitest.config.ts
cli/src/__tests__/company-delete.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
server/src/__tests__/productivity-review-service.test.ts` -> 3 files, 18
tests passed
- `pnpm exec vitest run --config vitest.config.ts
server/src/__tests__/issue-attachment-routes.test.ts` -> 1 file, 6 tests
passed
- `pnpm --filter @paperclipai/db typecheck && pnpm --filter
@paperclipai/shared typecheck && pnpm --filter @paperclipai/server
typecheck && pnpm --filter paperclipai typecheck`
- `pnpm --filter @paperclipai/server typecheck`
- `pnpm --filter @paperclipai/ui typecheck && pnpm --filter
@paperclipai/ui build`
## Risks
- Includes migrations `0073_shiny_salo.sql` and
`0074_striped_genesis.sql`; merge ordering matters if another PR adds
migrations first.
- This is intentionally backend-only apart from fixture/test updates
forced by shared type correctness; UI affordances from PR #4692 are not
present here and should land in separate UI slices.
- The worktree install emitted plugin SDK bin-link warnings for unbuilt
plugin packages, but the targeted tests and package typechecks completed
successfully.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected; check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex, GPT-5 coding agent, tool-enabled terminal/GitHub
workflow. Exact runtime context window was not exposed by the harness.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-04-28 16:46:45 -05:00
} ;
}
function choosePrimaryTrigger ( input : {
noComment : boolean ;
longActive : boolean ;
highChurn : boolean ;
} ) : ProductivityReviewTrigger | null {
if ( input . noComment ) return "no_comment_streak" ;
if ( input . highChurn ) return "high_churn" ;
if ( input . longActive ) return "long_active_duration" ;
return null ;
}
function isSoftStopTrigger ( trigger : ProductivityReviewTrigger ) {
return trigger === "no_comment_streak" || trigger === "high_churn" ;
}
function formatTrigger ( trigger : ProductivityReviewTrigger ) {
if ( trigger === "no_comment_streak" ) return "No-comment streak" ;
if ( trigger === "high_churn" ) return "High churn" ;
return "Long active duration" ;
}
export function productivityReviewService ( db : Db , deps ? : { enqueueWakeup? : EnqueueWakeup } ) {
const issuesSvc = issueService ( db ) ;
const budgets = budgetService ( db ) ;
async function getCompanyIssuePrefix ( companyId : string ) {
return db
. select ( { issuePrefix : companies.issuePrefix } )
. from ( companies )
. where ( eq ( companies . id , companyId ) )
. then ( ( rows ) = > rows [ 0 ] ? . issuePrefix ? ? "PAP" ) ;
}
async function getAgent ( agentId : string ) {
return db
. select ( )
. from ( agents )
. where ( eq ( agents . id , agentId ) )
. then ( ( rows ) = > rows [ 0 ] ? ? null ) ;
}
function isAgentInvokable ( agent : AgentRow | null | undefined ) {
return Boolean ( agent && ! [ "paused" , "terminated" , "pending_approval" ] . includes ( agent . status ) ) ;
}
async function isProductivityReviewDescendant ( issue : Pick < IssueRow , "companyId" | "parentId" > ) {
let parentId = issue . parentId ;
let depth = 0 ;
while ( parentId && depth < MAX_PARENT_WALK_DEPTH ) {
const parent = await db
. select ( { id : issues.id , parentId : issues.parentId , originKind : issues.originKind } )
. from ( issues )
. where ( and ( eq ( issues . companyId , issue . companyId ) , eq ( issues . id , parentId ) ) )
. then ( ( rows ) = > rows [ 0 ] ? ? null ) ;
if ( ! parent ) return false ;
if ( parent . originKind === PRODUCTIVITY_REVIEW_ORIGIN_KIND ) return true ;
parentId = parent . parentId ;
depth += 1 ;
}
return false ;
}
async function findOpenProductivityReview ( companyId : string , sourceIssueId : string ) {
return db
. select ( )
. from ( issues )
. where (
and (
eq ( issues . companyId , companyId ) ,
eq ( issues . originKind , PRODUCTIVITY_REVIEW_ORIGIN_KIND ) ,
eq ( issues . originId , sourceIssueId ) ,
isNull ( issues . hiddenAt ) ,
notInArray ( issues . status , [ "done" , "cancelled" ] ) ,
) ,
)
. orderBy ( desc ( issues . updatedAt ) )
. limit ( 1 )
. then ( ( rows ) = > rows [ 0 ] ? ? null ) ;
}
async function findRecentResolvedProductivityReview (
companyId : string ,
sourceIssueId : string ,
thresholds : ProductivityReviewThresholds ,
now : Date ,
) {
const cutoff = new Date ( now . getTime ( ) - thresholds . resolvedSnoozeMs ) ;
return db
. select ( { id : issues.id , identifier : issues.identifier , status : issues.status , updatedAt : issues.updatedAt } )
. from ( issues )
. where (
and (
eq ( issues . companyId , companyId ) ,
eq ( issues . originKind , PRODUCTIVITY_REVIEW_ORIGIN_KIND ) ,
eq ( issues . originId , sourceIssueId ) ,
eq ( issues . status , "done" ) ,
gt ( issues . updatedAt , cutoff ) ,
) ,
)
. orderBy ( desc ( issues . updatedAt ) )
. limit ( 1 )
. then ( ( rows ) = > rows [ 0 ] ? ? null ) ;
}
[codex] Bound productivity review recovery loops (#4948)
## Thinking Path
> - Paperclip orchestrates AI agents for zero-human companies.
> - The heartbeat/productivity review subsystem detects when assigned
work is likely stuck or churning.
> - Productivity reviews are useful, but repeated reconciliation can
create noisy refresh comments or repeated review issues around the same
source issue.
> - That makes manager follow-up harder because the signal can get
buried under duplicate review activity.
> - This pull request bounds productivity review refreshes and creation
loops while preserving the existing escalation path.
> - The benefit is a quieter recovery loop that still surfaces stuck or
high-churn work for manager attention.
## What Changed
- Added refresh throttling for open productivity review issues,
including a one-hour default interval and a maximum of three refresh
comments per open review.
- Added a rolling 24-hour creation cap so completed/closed reviews
cannot immediately recreate review issues indefinitely for the same
source issue.
- Excluded cancelled productivity reviews from the creation cap so
manager cancellations do not silently suppress future legitimate
reviews.
- Preserved productivity review timestamps in deterministic test paths
and added targeted coverage for immediate refresh suppression, refresh
caps, creation caps, and cancelled-review exclusion.
## Verification
- `pnpm run preflight:workspace-links && pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- `pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- Greptile Review: 5/5 on commit
`bcf25832d0ffae25890b2ee7eed112d1c2d114fe` with review threads resolved.
- GitHub PR checks passed on the latest head: `policy`, `verify`, `e2e`,
`Greptile Review`, and `security/snyk (cryppadotta)`.
- Verified the branch is rebased onto `public-gh/master` with no
conflicts.
- Verified the diff does not include `pnpm-lock.yaml`, database schema
changes, or migrations.
## Risks
- Low-to-medium risk: this changes automation cadence for productivity
reviews. A truly stuck issue may receive fewer repeated refresh
comments, but the original review issue remains open and assigned for
manager action.
- No migration risk: this is server logic and tests only.
> Checked [`ROADMAP.md`](ROADMAP.md) for overlapping planned core work;
this is a targeted recovery-loop fix and does not add a new roadmap
feature.
## Model Used
- OpenAI Codex coding agent, GPT-5 model family, tool-using software
engineering mode. Exact context window is not exposed in this runtime.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots (not applicable; server-only change)
- [x] I have updated relevant documentation to reflect my changes (not
applicable; no user-facing docs or commands changed)
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-05-01 08:32:04 -05:00
async function countRecentProductivityReviews (
companyId : string ,
sourceIssueId : string ,
thresholds : ProductivityReviewThresholds ,
now : Date ,
) {
const cutoff = new Date ( now . getTime ( ) - thresholds . creationWindowMs ) ;
return db
. select ( { count : sql < number > ` count(*)::int ` } )
. from ( issues )
. where (
and (
eq ( issues . companyId , companyId ) ,
eq ( issues . originKind , PRODUCTIVITY_REVIEW_ORIGIN_KIND ) ,
eq ( issues . originId , sourceIssueId ) ,
isNull ( issues . hiddenAt ) ,
sql ` ${ issues . status } <> 'cancelled' ` ,
sql ` ${ issues . createdAt } >= ${ cutoff . toISOString ( ) } ::timestamptz ` ,
) ,
)
. then ( ( rows ) = > Number ( rows [ 0 ] ? . count ? ? 0 ) ) ;
}
async function getRefreshCommentState ( companyId : string , reviewIssueId : string ) {
return db
. select ( {
count : sql < number > ` count(*)::int ` ,
latestCreatedAt : sql < Date | null > ` max( ${ issueComments . createdAt } ) ` ,
} )
. from ( issueComments )
. where (
and (
eq ( issueComments . companyId , companyId ) ,
eq ( issueComments . issueId , reviewIssueId ) ,
sql ` ${ issueComments . body } like ${ ` ${ PRODUCTIVITY_REVIEW_REFRESH_COMMENT_PREFIX } % ` } ` ,
) ,
)
. then ( ( rows ) = > {
const row = rows [ 0 ] ;
return {
count : Number ( row ? . count ? ? 0 ) ,
latestCreatedAt : coerceDate ( row ? . latestCreatedAt ) ,
} ;
} ) ;
}
async function addRefreshComment (
reviewIssueId : string ,
body : string ,
generatedAt : Date ,
) {
const comment = await issuesSvc . addComment ( reviewIssueId , body , { } ) ;
await db
. update ( issueComments )
. set ( { createdAt : generatedAt , updatedAt : generatedAt } )
. where ( eq ( issueComments . id , comment . id ) ) ;
await db
. update ( issues )
. set ( { updatedAt : generatedAt } )
. where ( eq ( issues . id , reviewIssueId ) ) ;
return comment ;
}
[codex] Split backend control-plane QoL slice (#4700)
## Thinking Path
> - Paperclip is the control plane for autonomous AI companies, so
backend task ownership, recovery, review visibility, and company-scoped
limits need to stay enforceable without UI-only coupling.
> - Closed PR #4692 bundled those backend changes with UI workflow,
docs, skills, workflow, and lockfile churn.
> - PAP-2694 asks for a clean backend/control-plane slice from that
closed branch.
> - This branch starts from current `master` and mines only the `cli`,
`packages/db`, `packages/shared`, and `server` contracts/tests needed
for the backend behavior.
> - It explicitly excludes UI workflow/performance work,
`.github/workflows/pr.yml`, `pnpm-lock.yaml`, docs, skills,
package-script, adapter UI build-config, and perf fixture script
changes; the only UI files are fixture/test updates required by the
tightened shared `Company` contract.
> - The benefit is a smaller reviewable PR that preserves the
control-plane fixes while staying under Greptile s 100-file review
limit.
## What Changed
- Added company-scoped attachment-size limits through DB
schema/migrations, shared company portability contracts, CLI
import/export coverage, and server attachment upload enforcement.
- Added productivity review service/API behavior for no-comment streak,
long-active, and high-churn review issues, including request-depth
clamping and issue summary exposure.
- Hardened issue ownership and recovery/control-plane paths: peer-agent
mutation denial, issue tree pause/resume behavior, stranded recovery
origins, and related activity/test coverage.
- Preserved related backend contract updates for routine timestamp
variables and managed agent instruction bundles because they live in
shared/server contracts from the source branch.
- Addressed Greptile feedback by making `Company.attachmentMaxBytes`
non-optional, simplifying review request-depth clamping, fixing the
migration final newline, and enforcing the process-level attachment cap
as the final ceiling for uploads.
- Added minimal company fixtures needed for repo-wide typecheck/build
and kept the PR to 66 changed files with forbidden/non-slice paths
excluded.
## Verification
- `pnpm install --frozen-lockfile`
- `git diff --check origin/master..HEAD`
- `git diff --name-only origin/master..HEAD | wc -l` -> 66 files
- `git diff --name-only origin/master..HEAD -- .github/workflows/pr.yml
pnpm-lock.yaml package.json doc skills .agents scripts
packages/adapters` -> no output
- `pnpm exec vitest run --config vitest.config.ts
packages/shared/src/validators/issue.test.ts
packages/shared/src/routine-variables.test.ts
packages/shared/src/adapter-types.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
cli/src/__tests__/company.test.ts
server/src/__tests__/productivity-review-service.test.ts
server/src/__tests__/issue-tree-control-service.test.ts
server/src/__tests__/issue-tree-control-routes.test.ts
server/src/__tests__/issue-agent-mutation-ownership-routes.test.ts
server/src/__tests__/issue-attachment-routes.test.ts
server/src/__tests__/heartbeat-process-recovery.test.ts
server/src/__tests__/issues-service.test.ts` -> 12 files, 147 tests
passed
- `pnpm exec vitest run --config vitest.config.ts
cli/src/__tests__/company-delete.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
server/src/__tests__/productivity-review-service.test.ts` -> 3 files, 18
tests passed
- `pnpm exec vitest run --config vitest.config.ts
server/src/__tests__/issue-attachment-routes.test.ts` -> 1 file, 6 tests
passed
- `pnpm --filter @paperclipai/db typecheck && pnpm --filter
@paperclipai/shared typecheck && pnpm --filter @paperclipai/server
typecheck && pnpm --filter paperclipai typecheck`
- `pnpm --filter @paperclipai/server typecheck`
- `pnpm --filter @paperclipai/ui typecheck && pnpm --filter
@paperclipai/ui build`
## Risks
- Includes migrations `0073_shiny_salo.sql` and
`0074_striped_genesis.sql`; merge ordering matters if another PR adds
migrations first.
- This is intentionally backend-only apart from fixture/test updates
forced by shared type correctness; UI affordances from PR #4692 are not
present here and should land in separate UI slices.
- The worktree install emitted plugin SDK bin-link warnings for unbuilt
plugin packages, but the targeted tests and package typechecks completed
successfully.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected; check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex, GPT-5 coding agent, tool-enabled terminal/GitHub
workflow. Exact runtime context window was not exposed by the harness.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-04-28 16:46:45 -05:00
async function countIssueRunsSince ( companyId : string , agentId : string , issueId : string , since : Date ) {
return db
. select ( { count : sql < number > ` count(*)::int ` } )
. from ( heartbeatRuns )
. where (
and (
eq ( heartbeatRuns . companyId , companyId ) ,
eq ( heartbeatRuns . agentId , agentId ) ,
issueRunScopeSql ( issueId ) ,
sql ` coalesce( ${ heartbeatRuns . startedAt } , ${ heartbeatRuns . createdAt } ) >= ${ since . toISOString ( ) } ::timestamptz ` ,
) ,
)
. then ( ( rows ) = > rows [ 0 ] ? . count ? ? 0 ) ;
}
async function countIssueCommentsSince ( companyId : string , issueId : string , agentId : string , since? : Date ) {
return db
. select ( { count : sql < number > ` count(*)::int ` } )
. from ( issueComments )
. innerJoin ( heartbeatRuns , eq ( heartbeatRuns . id , issueComments . createdByRunId ) )
. where (
and (
eq ( issueComments . companyId , companyId ) ,
eq ( issueComments . issueId , issueId ) ,
eq ( issueComments . authorAgentId , agentId ) ,
eq ( heartbeatRuns . companyId , companyId ) ,
eq ( heartbeatRuns . agentId , agentId ) ,
issueRunScopeSql ( issueId ) ,
since ? sql ` ${ issueComments . createdAt } >= ${ since . toISOString ( ) } ::timestamptz ` : undefined ,
) ,
)
. then ( ( rows ) = > rows [ 0 ] ? . count ? ? 0 ) ;
}
async function collectEvidence (
sourceIssue : IssueRow ,
sourceAgent : AgentRow ,
thresholds : ProductivityReviewThresholds ,
now : Date ,
) : Promise < ProductivityReviewEvidence | null > {
const oneHourAgo = new Date ( now . getTime ( ) - 60 * 60 * 1000 ) ;
const sixHoursAgo = new Date ( now . getTime ( ) - 6 * 60 * 60 * 1000 ) ;
const latestRuns = await db
. select ( )
. from ( heartbeatRuns )
. where (
and (
eq ( heartbeatRuns . companyId , sourceIssue . companyId ) ,
eq ( heartbeatRuns . agentId , sourceAgent . id ) ,
issueRunScopeSql ( sourceIssue . id ) ,
) ,
)
. orderBy ( desc ( heartbeatRuns . createdAt ) , desc ( heartbeatRuns . id ) )
. limit ( MAX_RUNS_FOR_STREAK ) ;
const runIds = latestRuns . map ( ( run ) = > run . id ) ;
const commentRunIds = new Set < string > ( ) ;
if ( runIds . length > 0 ) {
const commentRows = await db
. select ( { createdByRunId : issueComments.createdByRunId } )
. from ( issueComments )
. where (
and (
eq ( issueComments . companyId , sourceIssue . companyId ) ,
eq ( issueComments . issueId , sourceIssue . id ) ,
inArray ( issueComments . createdByRunId , runIds ) ,
) ,
) ;
for ( const row of commentRows ) {
if ( row . createdByRunId ) commentRunIds . add ( row . createdByRunId ) ;
}
}
const terminalRuns = latestRuns . filter ( ( run ) = >
TERMINAL_RUN_STATUSES . includes ( run . status as ( typeof TERMINAL_RUN_STATUSES ) [ number ] ) ,
) ;
let noCommentStreak = 0 ;
for ( const run of terminalRuns ) {
if ( commentRunIds . has ( run . id ) ) break ;
noCommentStreak += 1 ;
}
const [
runCountLastHour ,
runCountLastSixHours ,
assigneeRunCommentCount ,
assigneeRunCommentCountLastHour ,
assigneeRunCommentCountLastSixHours ,
latestComments ,
costRow ,
] = await Promise . all ( [
countIssueRunsSince ( sourceIssue . companyId , sourceAgent . id , sourceIssue . id , oneHourAgo ) ,
countIssueRunsSince ( sourceIssue . companyId , sourceAgent . id , sourceIssue . id , sixHoursAgo ) ,
countIssueCommentsSince ( sourceIssue . companyId , sourceIssue . id , sourceAgent . id ) ,
countIssueCommentsSince ( sourceIssue . companyId , sourceIssue . id , sourceAgent . id , oneHourAgo ) ,
countIssueCommentsSince ( sourceIssue . companyId , sourceIssue . id , sourceAgent . id , sixHoursAgo ) ,
db
. select ( { comment : issueComments } )
. from ( issueComments )
. innerJoin ( heartbeatRuns , eq ( heartbeatRuns . id , issueComments . createdByRunId ) )
. where (
and (
eq ( issueComments . companyId , sourceIssue . companyId ) ,
eq ( issueComments . issueId , sourceIssue . id ) ,
eq ( issueComments . authorAgentId , sourceAgent . id ) ,
eq ( heartbeatRuns . companyId , sourceIssue . companyId ) ,
eq ( heartbeatRuns . agentId , sourceAgent . id ) ,
issueRunScopeSql ( sourceIssue . id ) ,
) ,
)
. orderBy ( desc ( issueComments . createdAt ) , desc ( issueComments . id ) )
. limit ( 5 )
. then ( ( rows ) = > rows . map ( ( row ) = > row . comment ) ) ,
db
. select ( { costCents : sql < number > ` coalesce(sum( ${ costEvents . costCents } ), 0)::int ` } )
. from ( costEvents )
. where ( and ( eq ( costEvents . companyId , sourceIssue . companyId ) , eq ( costEvents . issueId , sourceIssue . id ) ) )
. then ( ( rows ) = > rows [ 0 ] ? ? { costCents : 0 } ) ,
] ) ;
const activeRunCount = latestRuns . filter ( ( run ) = >
ACTIVE_RUN_STATUSES . includes ( run . status as ( typeof ACTIVE_RUN_STATUSES ) [ number ] ) ,
) . length ;
const activeStartedAt = sourceIssue . startedAt ? ? sourceIssue . executionLockedAt ? ? null ;
const elapsedMs = sourceIssue . status === "in_progress" && activeStartedAt
? Math . max ( 0 , now . getTime ( ) - activeStartedAt . getTime ( ) )
: null ;
const noComment = noCommentStreak >= thresholds . noCommentStreakRuns ;
const longActive = elapsedMs !== null && elapsedMs >= thresholds . longActiveMs ;
const highChurn =
runCountLastHour >= thresholds . highChurnHourly ||
assigneeRunCommentCountLastHour >= thresholds . highChurnHourly ||
runCountLastSixHours >= thresholds . highChurnSixHours ||
assigneeRunCommentCountLastSixHours >= thresholds . highChurnSixHours ;
const trigger = choosePrimaryTrigger ( { noComment , longActive , highChurn } ) ;
if ( ! trigger ) return null ;
const triggerReasons : string [ ] = [ ] ;
if ( noComment ) triggerReasons . push ( ` ${ noCommentStreak } consecutive completed issue-linked runs had no run-created issue comment ` ) ;
if ( longActive ) triggerReasons . push ( ` current active episode has lasted ${ msToHuman ( elapsedMs ) } ` ) ;
if ( highChurn ) {
triggerReasons . push (
` ${ runCountLastHour } runs/ ${ assigneeRunCommentCountLastHour } assignee-run comments in 1h; ${ runCountLastSixHours } runs/ ${ assigneeRunCommentCountLastSixHours } assignee-run comments in 6h ` ,
) ;
}
return {
trigger ,
triggerReasons ,
sourceIssue ,
sourceAgent ,
noCommentStreak ,
totalRunCount : latestRuns.length ,
terminalRunCount : terminalRuns.length ,
activeRunCount ,
runCountLastHour ,
runCountLastSixHours ,
commentCount : assigneeRunCommentCount ,
commentCountLastHour : assigneeRunCommentCountLastHour ,
commentCountLastSixHours : assigneeRunCommentCountLastSixHours ,
elapsedMs ,
latestRuns : latestRuns.slice ( 0 , 5 ) ,
latestComments ,
costCents : costRow.costCents ,
usageSamples : latestRuns
. filter ( ( run ) = > run . usageJson )
. slice ( 0 , 3 )
. map ( ( run ) = > ( { runId : run.id , usageJson : run.usageJson ? ? null } ) ) ,
nextAction : latestRuns.find ( ( run ) = > run . nextAction ) ? . nextAction ? ? null ,
thresholds ,
generatedAt : now ,
} ;
}
async function resolveReviewOwnerAgentId ( sourceIssue : IssueRow , sourceAgent : AgentRow ) {
const candidateIds : string [ ] = [ ] ;
if ( sourceAgent . reportsTo ) candidateIds . push ( sourceAgent . reportsTo ) ;
if ( sourceIssue . createdByAgentId ) candidateIds . push ( sourceIssue . createdByAgentId ) ;
if ( sourceIssue . projectId ) {
const project = await db
. select ( { leadAgentId : projects.leadAgentId } )
. from ( projects )
. where ( and ( eq ( projects . companyId , sourceIssue . companyId ) , eq ( projects . id , sourceIssue . projectId ) ) )
. then ( ( rows ) = > rows [ 0 ] ? ? null ) ;
if ( project ? . leadAgentId ) candidateIds . push ( project . leadAgentId ) ;
}
const roleCandidates = await db
. select ( { id : agents.id } )
. from ( agents )
. where ( and ( eq ( agents . companyId , sourceIssue . companyId ) , inArray ( agents . role , [ "cto" , "ceo" ] ) ) )
. orderBy ( sql ` case when ${ agents . role } = 'cto' then 0 else 1 end ` , asc ( agents . createdAt ) , asc ( agents . id ) ) ;
candidateIds . push ( . . . roleCandidates . map ( ( agent ) = > agent . id ) ) ;
const seen = new Set < string > ( ) ;
for ( const agentId of candidateIds ) {
if ( seen . has ( agentId ) ) continue ;
seen . add ( agentId ) ;
const candidate = await getAgent ( agentId ) ;
if ( ! candidate || candidate . companyId !== sourceIssue . companyId || ! isAgentInvokable ( candidate ) ) continue ;
const budgetBlock = await budgets . getInvocationBlock ( sourceIssue . companyId , candidate . id , {
issueId : sourceIssue.id ,
projectId : sourceIssue.projectId ? ? null ,
} ) ;
if ( ! budgetBlock ) return candidate . id ;
}
return null ;
}
function buildReviewMarkdown ( evidence : ProductivityReviewEvidence , prefix : string ) {
const latestRuns = evidence . latestRuns . length > 0
? evidence . latestRuns . map ( ( run ) = >
` - ${ runUiLink ( run , prefix ) } \` ${ run . status } \` liveness \` ${ run . livenessState ? ? "unknown" } \` , created ${ run . createdAt . toISOString ( ) } ${ run . nextAction ? ` , next action: ${ truncateInline ( run . nextAction , 160 ) } ` : "" } ` ,
) . join ( "\n" )
: "- none" ;
const latestComments = evidence . latestComments . length > 0
? evidence . latestComments . map ( ( comment ) = >
` - ${ comment . createdAt . toISOString ( ) } ${ comment . createdByRunId ? ` run \` ${ comment . createdByRunId } \` ` : "" } : ${ truncateInline ( comment . body ) } ` ,
) . join ( "\n" )
: "- none" ;
const usage = evidence . usageSamples . length > 0
? evidence . usageSamples . map ( ( sample ) = > ` - \` ${ sample . runId } \` : \` ${ JSON . stringify ( sample . usageJson ) . slice ( 0 , 500 ) } \` ` ) . join ( "\n" )
: "- no usage payloads on sampled runs" ;
return [
"Paperclip detected an unusual productivity/progression pattern on an assigned issue." ,
"" ,
"## Source" ,
"" ,
` - Source issue: ${ issueUiLink ( evidence . sourceIssue , prefix ) } ` ,
` - Assigned agent: ${ evidence . sourceAgent . name } ( ${ evidence . sourceAgent . role } ) ` ,
` - Primary trigger: \` ${ evidence . trigger } \` ( ${ formatTrigger ( evidence . trigger ) } ) ` ,
` - Trigger reasons: ${ evidence . triggerReasons . join ( "; " ) } ` ,
` - Generated at: ${ evidence . generatedAt . toISOString ( ) } ` ,
"" ,
"## Evidence" ,
"" ,
` - Total sampled issue-linked runs: ${ evidence . totalRunCount } ` ,
` - Terminal sampled runs: ${ evidence . terminalRunCount } ` ,
` - Active queued/running/scheduled runs: ${ evidence . activeRunCount } ` ,
` - No-comment completed-run streak: ${ evidence . noCommentStreak } ` ,
` - Current active elapsed time: ${ msToHuman ( evidence . elapsedMs ) } ` ,
` - Runs in rolling windows: ${ evidence . runCountLastHour } /1h, ${ evidence . runCountLastSixHours } /6h ` ,
` - Assignee run-linked comments total/window: ${ evidence . commentCount } total, ${ evidence . commentCountLastHour } /1h, ${ evidence . commentCountLastSixHours } /6h ` ,
` - Cost events total: ${ evidence . costCents } cents ` ,
` - Current next action: ${ evidence . nextAction ? truncateInline ( evidence . nextAction , 500 ) : "none recorded" } ` ,
"" ,
"## Thresholds" ,
"" ,
` - No-comment streak: ${ evidence . thresholds . noCommentStreakRuns } completed runs ` ,
` - Long active duration: ${ msToHuman ( evidence . thresholds . longActiveMs ) } ` ,
` - High churn: ${ evidence . thresholds . highChurnHourly } /1h or ${ evidence . thresholds . highChurnSixHours } /6h runs/assignee-run comments ` ,
` - Resolved-review snooze: ${ msToHuman ( evidence . thresholds . resolvedSnoozeMs ) } ` ,
"" ,
"## Latest Runs" ,
"" ,
latestRuns ,
"" ,
"## Latest Assignee Run Comments" ,
"" ,
latestComments ,
"" ,
"## Usage Samples" ,
"" ,
usage ,
"" ,
"## Manager Decision" ,
"" ,
"- Close as productive if this pattern is expected." ,
"- Continue with a snooze window if the current work should keep running without repeat review spam." ,
"- Request decomposition, reroute, block with an unblock owner, or stop/cancel the source work if the work is inefficient." ,
] . join ( "\n" ) ;
}
function buildRefreshComment ( evidence : ProductivityReviewEvidence , prefix : string ) {
return [
"Productivity review evidence refreshed." ,
"" ,
` - Source issue: ${ issueUiLink ( evidence . sourceIssue , prefix ) } ` ,
` - Trigger: \` ${ evidence . trigger } \` ( ${ formatTrigger ( evidence . trigger ) } ) ` ,
` - Reasons: ${ evidence . triggerReasons . join ( "; " ) } ` ,
` - No-comment streak: ${ evidence . noCommentStreak } ` ,
` - Runs/assignee comments: ${ evidence . runCountLastHour } / ${ evidence . commentCountLastHour } in 1h, ${ evidence . runCountLastSixHours } / ${ evidence . commentCountLastSixHours } in 6h ` ,
` - Next action: ${ evidence . nextAction ? truncateInline ( evidence . nextAction , 300 ) : "none recorded" } ` ,
] . join ( "\n" ) ;
}
async function createOrUpdateReview (
evidence : ProductivityReviewEvidence ,
[codex] Bound productivity review recovery loops (#4948)
## Thinking Path
> - Paperclip orchestrates AI agents for zero-human companies.
> - The heartbeat/productivity review subsystem detects when assigned
work is likely stuck or churning.
> - Productivity reviews are useful, but repeated reconciliation can
create noisy refresh comments or repeated review issues around the same
source issue.
> - That makes manager follow-up harder because the signal can get
buried under duplicate review activity.
> - This pull request bounds productivity review refreshes and creation
loops while preserving the existing escalation path.
> - The benefit is a quieter recovery loop that still surfaces stuck or
high-churn work for manager attention.
## What Changed
- Added refresh throttling for open productivity review issues,
including a one-hour default interval and a maximum of three refresh
comments per open review.
- Added a rolling 24-hour creation cap so completed/closed reviews
cannot immediately recreate review issues indefinitely for the same
source issue.
- Excluded cancelled productivity reviews from the creation cap so
manager cancellations do not silently suppress future legitimate
reviews.
- Preserved productivity review timestamps in deterministic test paths
and added targeted coverage for immediate refresh suppression, refresh
caps, creation caps, and cancelled-review exclusion.
## Verification
- `pnpm run preflight:workspace-links && pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- `pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- Greptile Review: 5/5 on commit
`bcf25832d0ffae25890b2ee7eed112d1c2d114fe` with review threads resolved.
- GitHub PR checks passed on the latest head: `policy`, `verify`, `e2e`,
`Greptile Review`, and `security/snyk (cryppadotta)`.
- Verified the branch is rebased onto `public-gh/master` with no
conflicts.
- Verified the diff does not include `pnpm-lock.yaml`, database schema
changes, or migrations.
## Risks
- Low-to-medium risk: this changes automation cadence for productivity
reviews. A truly stuck issue may receive fewer repeated refresh
comments, but the original review issue remains open and assigned for
manager action.
- No migration risk: this is server logic and tests only.
> Checked [`ROADMAP.md`](ROADMAP.md) for overlapping planned core work;
this is a targeted recovery-loop fix and does not add a new roadmap
feature.
## Model Used
- OpenAI Codex coding agent, GPT-5 model family, tool-using software
engineering mode. Exact context window is not exposed in this runtime.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots (not applicable; server-only change)
- [x] I have updated relevant documentation to reflect my changes (not
applicable; no user-facing docs or commands changed)
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-05-01 08:32:04 -05:00
opts : { prefix : string ; thresholds : ProductivityReviewThresholds } ,
[codex] Split backend control-plane QoL slice (#4700)
## Thinking Path
> - Paperclip is the control plane for autonomous AI companies, so
backend task ownership, recovery, review visibility, and company-scoped
limits need to stay enforceable without UI-only coupling.
> - Closed PR #4692 bundled those backend changes with UI workflow,
docs, skills, workflow, and lockfile churn.
> - PAP-2694 asks for a clean backend/control-plane slice from that
closed branch.
> - This branch starts from current `master` and mines only the `cli`,
`packages/db`, `packages/shared`, and `server` contracts/tests needed
for the backend behavior.
> - It explicitly excludes UI workflow/performance work,
`.github/workflows/pr.yml`, `pnpm-lock.yaml`, docs, skills,
package-script, adapter UI build-config, and perf fixture script
changes; the only UI files are fixture/test updates required by the
tightened shared `Company` contract.
> - The benefit is a smaller reviewable PR that preserves the
control-plane fixes while staying under Greptile s 100-file review
limit.
## What Changed
- Added company-scoped attachment-size limits through DB
schema/migrations, shared company portability contracts, CLI
import/export coverage, and server attachment upload enforcement.
- Added productivity review service/API behavior for no-comment streak,
long-active, and high-churn review issues, including request-depth
clamping and issue summary exposure.
- Hardened issue ownership and recovery/control-plane paths: peer-agent
mutation denial, issue tree pause/resume behavior, stranded recovery
origins, and related activity/test coverage.
- Preserved related backend contract updates for routine timestamp
variables and managed agent instruction bundles because they live in
shared/server contracts from the source branch.
- Addressed Greptile feedback by making `Company.attachmentMaxBytes`
non-optional, simplifying review request-depth clamping, fixing the
migration final newline, and enforcing the process-level attachment cap
as the final ceiling for uploads.
- Added minimal company fixtures needed for repo-wide typecheck/build
and kept the PR to 66 changed files with forbidden/non-slice paths
excluded.
## Verification
- `pnpm install --frozen-lockfile`
- `git diff --check origin/master..HEAD`
- `git diff --name-only origin/master..HEAD | wc -l` -> 66 files
- `git diff --name-only origin/master..HEAD -- .github/workflows/pr.yml
pnpm-lock.yaml package.json doc skills .agents scripts
packages/adapters` -> no output
- `pnpm exec vitest run --config vitest.config.ts
packages/shared/src/validators/issue.test.ts
packages/shared/src/routine-variables.test.ts
packages/shared/src/adapter-types.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
cli/src/__tests__/company.test.ts
server/src/__tests__/productivity-review-service.test.ts
server/src/__tests__/issue-tree-control-service.test.ts
server/src/__tests__/issue-tree-control-routes.test.ts
server/src/__tests__/issue-agent-mutation-ownership-routes.test.ts
server/src/__tests__/issue-attachment-routes.test.ts
server/src/__tests__/heartbeat-process-recovery.test.ts
server/src/__tests__/issues-service.test.ts` -> 12 files, 147 tests
passed
- `pnpm exec vitest run --config vitest.config.ts
cli/src/__tests__/company-delete.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
server/src/__tests__/productivity-review-service.test.ts` -> 3 files, 18
tests passed
- `pnpm exec vitest run --config vitest.config.ts
server/src/__tests__/issue-attachment-routes.test.ts` -> 1 file, 6 tests
passed
- `pnpm --filter @paperclipai/db typecheck && pnpm --filter
@paperclipai/shared typecheck && pnpm --filter @paperclipai/server
typecheck && pnpm --filter paperclipai typecheck`
- `pnpm --filter @paperclipai/server typecheck`
- `pnpm --filter @paperclipai/ui typecheck && pnpm --filter
@paperclipai/ui build`
## Risks
- Includes migrations `0073_shiny_salo.sql` and
`0074_striped_genesis.sql`; merge ordering matters if another PR adds
migrations first.
- This is intentionally backend-only apart from fixture/test updates
forced by shared type correctness; UI affordances from PR #4692 are not
present here and should land in separate UI slices.
- The worktree install emitted plugin SDK bin-link warnings for unbuilt
plugin packages, but the targeted tests and package typechecks completed
successfully.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected; check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex, GPT-5 coding agent, tool-enabled terminal/GitHub
workflow. Exact runtime context window was not exposed by the harness.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-04-28 16:46:45 -05:00
) {
const existing = await findOpenProductivityReview ( evidence . sourceIssue . companyId , evidence . sourceIssue . id ) ;
if ( existing ) {
[codex] Bound productivity review recovery loops (#4948)
## Thinking Path
> - Paperclip orchestrates AI agents for zero-human companies.
> - The heartbeat/productivity review subsystem detects when assigned
work is likely stuck or churning.
> - Productivity reviews are useful, but repeated reconciliation can
create noisy refresh comments or repeated review issues around the same
source issue.
> - That makes manager follow-up harder because the signal can get
buried under duplicate review activity.
> - This pull request bounds productivity review refreshes and creation
loops while preserving the existing escalation path.
> - The benefit is a quieter recovery loop that still surfaces stuck or
high-churn work for manager attention.
## What Changed
- Added refresh throttling for open productivity review issues,
including a one-hour default interval and a maximum of three refresh
comments per open review.
- Added a rolling 24-hour creation cap so completed/closed reviews
cannot immediately recreate review issues indefinitely for the same
source issue.
- Excluded cancelled productivity reviews from the creation cap so
manager cancellations do not silently suppress future legitimate
reviews.
- Preserved productivity review timestamps in deterministic test paths
and added targeted coverage for immediate refresh suppression, refresh
caps, creation caps, and cancelled-review exclusion.
## Verification
- `pnpm run preflight:workspace-links && pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- `pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- Greptile Review: 5/5 on commit
`bcf25832d0ffae25890b2ee7eed112d1c2d114fe` with review threads resolved.
- GitHub PR checks passed on the latest head: `policy`, `verify`, `e2e`,
`Greptile Review`, and `security/snyk (cryppadotta)`.
- Verified the branch is rebased onto `public-gh/master` with no
conflicts.
- Verified the diff does not include `pnpm-lock.yaml`, database schema
changes, or migrations.
## Risks
- Low-to-medium risk: this changes automation cadence for productivity
reviews. A truly stuck issue may receive fewer repeated refresh
comments, but the original review issue remains open and assigned for
manager action.
- No migration risk: this is server logic and tests only.
> Checked [`ROADMAP.md`](ROADMAP.md) for overlapping planned core work;
this is a targeted recovery-loop fix and does not add a new roadmap
feature.
## Model Used
- OpenAI Codex coding agent, GPT-5 model family, tool-using software
engineering mode. Exact context window is not exposed in this runtime.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots (not applicable; server-only change)
- [x] I have updated relevant documentation to reflect my changes (not
applicable; no user-facing docs or commands changed)
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-05-01 08:32:04 -05:00
const refreshState = await getRefreshCommentState ( evidence . sourceIssue . companyId , existing . id ) ;
const lastRefreshOrCreationAt = refreshState . latestCreatedAt ? ? existing . createdAt ;
if (
refreshState . count >= opts . thresholds . maxRefreshComments ||
evidence . generatedAt . getTime ( ) - lastRefreshOrCreationAt . getTime ( ) < opts . thresholds . refreshIntervalMs
) {
return { kind : "existing" as const , reviewIssueId : existing.id } ;
}
await addRefreshComment ( existing . id , buildRefreshComment ( evidence , opts . prefix ) , evidence . generatedAt ) ;
[codex] Split backend control-plane QoL slice (#4700)
## Thinking Path
> - Paperclip is the control plane for autonomous AI companies, so
backend task ownership, recovery, review visibility, and company-scoped
limits need to stay enforceable without UI-only coupling.
> - Closed PR #4692 bundled those backend changes with UI workflow,
docs, skills, workflow, and lockfile churn.
> - PAP-2694 asks for a clean backend/control-plane slice from that
closed branch.
> - This branch starts from current `master` and mines only the `cli`,
`packages/db`, `packages/shared`, and `server` contracts/tests needed
for the backend behavior.
> - It explicitly excludes UI workflow/performance work,
`.github/workflows/pr.yml`, `pnpm-lock.yaml`, docs, skills,
package-script, adapter UI build-config, and perf fixture script
changes; the only UI files are fixture/test updates required by the
tightened shared `Company` contract.
> - The benefit is a smaller reviewable PR that preserves the
control-plane fixes while staying under Greptile s 100-file review
limit.
## What Changed
- Added company-scoped attachment-size limits through DB
schema/migrations, shared company portability contracts, CLI
import/export coverage, and server attachment upload enforcement.
- Added productivity review service/API behavior for no-comment streak,
long-active, and high-churn review issues, including request-depth
clamping and issue summary exposure.
- Hardened issue ownership and recovery/control-plane paths: peer-agent
mutation denial, issue tree pause/resume behavior, stranded recovery
origins, and related activity/test coverage.
- Preserved related backend contract updates for routine timestamp
variables and managed agent instruction bundles because they live in
shared/server contracts from the source branch.
- Addressed Greptile feedback by making `Company.attachmentMaxBytes`
non-optional, simplifying review request-depth clamping, fixing the
migration final newline, and enforcing the process-level attachment cap
as the final ceiling for uploads.
- Added minimal company fixtures needed for repo-wide typecheck/build
and kept the PR to 66 changed files with forbidden/non-slice paths
excluded.
## Verification
- `pnpm install --frozen-lockfile`
- `git diff --check origin/master..HEAD`
- `git diff --name-only origin/master..HEAD | wc -l` -> 66 files
- `git diff --name-only origin/master..HEAD -- .github/workflows/pr.yml
pnpm-lock.yaml package.json doc skills .agents scripts
packages/adapters` -> no output
- `pnpm exec vitest run --config vitest.config.ts
packages/shared/src/validators/issue.test.ts
packages/shared/src/routine-variables.test.ts
packages/shared/src/adapter-types.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
cli/src/__tests__/company.test.ts
server/src/__tests__/productivity-review-service.test.ts
server/src/__tests__/issue-tree-control-service.test.ts
server/src/__tests__/issue-tree-control-routes.test.ts
server/src/__tests__/issue-agent-mutation-ownership-routes.test.ts
server/src/__tests__/issue-attachment-routes.test.ts
server/src/__tests__/heartbeat-process-recovery.test.ts
server/src/__tests__/issues-service.test.ts` -> 12 files, 147 tests
passed
- `pnpm exec vitest run --config vitest.config.ts
cli/src/__tests__/company-delete.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
server/src/__tests__/productivity-review-service.test.ts` -> 3 files, 18
tests passed
- `pnpm exec vitest run --config vitest.config.ts
server/src/__tests__/issue-attachment-routes.test.ts` -> 1 file, 6 tests
passed
- `pnpm --filter @paperclipai/db typecheck && pnpm --filter
@paperclipai/shared typecheck && pnpm --filter @paperclipai/server
typecheck && pnpm --filter paperclipai typecheck`
- `pnpm --filter @paperclipai/server typecheck`
- `pnpm --filter @paperclipai/ui typecheck && pnpm --filter
@paperclipai/ui build`
## Risks
- Includes migrations `0073_shiny_salo.sql` and
`0074_striped_genesis.sql`; merge ordering matters if another PR adds
migrations first.
- This is intentionally backend-only apart from fixture/test updates
forced by shared type correctness; UI affordances from PR #4692 are not
present here and should land in separate UI slices.
- The worktree install emitted plugin SDK bin-link warnings for unbuilt
plugin packages, but the targeted tests and package typechecks completed
successfully.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected; check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex, GPT-5 coding agent, tool-enabled terminal/GitHub
workflow. Exact runtime context window was not exposed by the harness.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-04-28 16:46:45 -05:00
await logActivity ( db , {
companyId : evidence.sourceIssue.companyId ,
actorType : "system" ,
actorId : "system" ,
action : "issue.productivity_review_updated" ,
entityType : "issue" ,
entityId : existing.id ,
agentId : existing.assigneeAgentId ,
details : {
source : "productivity_review.reconcile" ,
sourceIssueId : evidence.sourceIssue.id ,
trigger : evidence.trigger ,
noCommentStreak : evidence.noCommentStreak ,
runCountLastHour : evidence.runCountLastHour ,
commentCountLastHour : evidence.commentCountLastHour ,
} ,
} ) ;
return { kind : "updated" as const , reviewIssueId : existing.id } ;
}
[codex] Bound productivity review recovery loops (#4948)
## Thinking Path
> - Paperclip orchestrates AI agents for zero-human companies.
> - The heartbeat/productivity review subsystem detects when assigned
work is likely stuck or churning.
> - Productivity reviews are useful, but repeated reconciliation can
create noisy refresh comments or repeated review issues around the same
source issue.
> - That makes manager follow-up harder because the signal can get
buried under duplicate review activity.
> - This pull request bounds productivity review refreshes and creation
loops while preserving the existing escalation path.
> - The benefit is a quieter recovery loop that still surfaces stuck or
high-churn work for manager attention.
## What Changed
- Added refresh throttling for open productivity review issues,
including a one-hour default interval and a maximum of three refresh
comments per open review.
- Added a rolling 24-hour creation cap so completed/closed reviews
cannot immediately recreate review issues indefinitely for the same
source issue.
- Excluded cancelled productivity reviews from the creation cap so
manager cancellations do not silently suppress future legitimate
reviews.
- Preserved productivity review timestamps in deterministic test paths
and added targeted coverage for immediate refresh suppression, refresh
caps, creation caps, and cancelled-review exclusion.
## Verification
- `pnpm run preflight:workspace-links && pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- `pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- Greptile Review: 5/5 on commit
`bcf25832d0ffae25890b2ee7eed112d1c2d114fe` with review threads resolved.
- GitHub PR checks passed on the latest head: `policy`, `verify`, `e2e`,
`Greptile Review`, and `security/snyk (cryppadotta)`.
- Verified the branch is rebased onto `public-gh/master` with no
conflicts.
- Verified the diff does not include `pnpm-lock.yaml`, database schema
changes, or migrations.
## Risks
- Low-to-medium risk: this changes automation cadence for productivity
reviews. A truly stuck issue may receive fewer repeated refresh
comments, but the original review issue remains open and assigned for
manager action.
- No migration risk: this is server logic and tests only.
> Checked [`ROADMAP.md`](ROADMAP.md) for overlapping planned core work;
this is a targeted recovery-loop fix and does not add a new roadmap
feature.
## Model Used
- OpenAI Codex coding agent, GPT-5 model family, tool-using software
engineering mode. Exact context window is not exposed in this runtime.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots (not applicable; server-only change)
- [x] I have updated relevant documentation to reflect my changes (not
applicable; no user-facing docs or commands changed)
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-05-01 08:32:04 -05:00
const recentCreationCount = await countRecentProductivityReviews (
evidence . sourceIssue . companyId ,
evidence . sourceIssue . id ,
opts . thresholds ,
evidence . generatedAt ,
) ;
if ( recentCreationCount >= opts . thresholds . maxCreationsPerWindow ) {
return { kind : "creation_capped" as const , reviewIssueId : null } ;
}
[codex] Split backend control-plane QoL slice (#4700)
## Thinking Path
> - Paperclip is the control plane for autonomous AI companies, so
backend task ownership, recovery, review visibility, and company-scoped
limits need to stay enforceable without UI-only coupling.
> - Closed PR #4692 bundled those backend changes with UI workflow,
docs, skills, workflow, and lockfile churn.
> - PAP-2694 asks for a clean backend/control-plane slice from that
closed branch.
> - This branch starts from current `master` and mines only the `cli`,
`packages/db`, `packages/shared`, and `server` contracts/tests needed
for the backend behavior.
> - It explicitly excludes UI workflow/performance work,
`.github/workflows/pr.yml`, `pnpm-lock.yaml`, docs, skills,
package-script, adapter UI build-config, and perf fixture script
changes; the only UI files are fixture/test updates required by the
tightened shared `Company` contract.
> - The benefit is a smaller reviewable PR that preserves the
control-plane fixes while staying under Greptile s 100-file review
limit.
## What Changed
- Added company-scoped attachment-size limits through DB
schema/migrations, shared company portability contracts, CLI
import/export coverage, and server attachment upload enforcement.
- Added productivity review service/API behavior for no-comment streak,
long-active, and high-churn review issues, including request-depth
clamping and issue summary exposure.
- Hardened issue ownership and recovery/control-plane paths: peer-agent
mutation denial, issue tree pause/resume behavior, stranded recovery
origins, and related activity/test coverage.
- Preserved related backend contract updates for routine timestamp
variables and managed agent instruction bundles because they live in
shared/server contracts from the source branch.
- Addressed Greptile feedback by making `Company.attachmentMaxBytes`
non-optional, simplifying review request-depth clamping, fixing the
migration final newline, and enforcing the process-level attachment cap
as the final ceiling for uploads.
- Added minimal company fixtures needed for repo-wide typecheck/build
and kept the PR to 66 changed files with forbidden/non-slice paths
excluded.
## Verification
- `pnpm install --frozen-lockfile`
- `git diff --check origin/master..HEAD`
- `git diff --name-only origin/master..HEAD | wc -l` -> 66 files
- `git diff --name-only origin/master..HEAD -- .github/workflows/pr.yml
pnpm-lock.yaml package.json doc skills .agents scripts
packages/adapters` -> no output
- `pnpm exec vitest run --config vitest.config.ts
packages/shared/src/validators/issue.test.ts
packages/shared/src/routine-variables.test.ts
packages/shared/src/adapter-types.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
cli/src/__tests__/company.test.ts
server/src/__tests__/productivity-review-service.test.ts
server/src/__tests__/issue-tree-control-service.test.ts
server/src/__tests__/issue-tree-control-routes.test.ts
server/src/__tests__/issue-agent-mutation-ownership-routes.test.ts
server/src/__tests__/issue-attachment-routes.test.ts
server/src/__tests__/heartbeat-process-recovery.test.ts
server/src/__tests__/issues-service.test.ts` -> 12 files, 147 tests
passed
- `pnpm exec vitest run --config vitest.config.ts
cli/src/__tests__/company-delete.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
server/src/__tests__/productivity-review-service.test.ts` -> 3 files, 18
tests passed
- `pnpm exec vitest run --config vitest.config.ts
server/src/__tests__/issue-attachment-routes.test.ts` -> 1 file, 6 tests
passed
- `pnpm --filter @paperclipai/db typecheck && pnpm --filter
@paperclipai/shared typecheck && pnpm --filter @paperclipai/server
typecheck && pnpm --filter paperclipai typecheck`
- `pnpm --filter @paperclipai/server typecheck`
- `pnpm --filter @paperclipai/ui typecheck && pnpm --filter
@paperclipai/ui build`
## Risks
- Includes migrations `0073_shiny_salo.sql` and
`0074_striped_genesis.sql`; merge ordering matters if another PR adds
migrations first.
- This is intentionally backend-only apart from fixture/test updates
forced by shared type correctness; UI affordances from PR #4692 are not
present here and should land in separate UI slices.
- The worktree install emitted plugin SDK bin-link warnings for unbuilt
plugin packages, but the targeted tests and package typechecks completed
successfully.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected; check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex, GPT-5 coding agent, tool-enabled terminal/GitHub
workflow. Exact runtime context window was not exposed by the harness.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-04-28 16:46:45 -05:00
const ownerAgentId = await resolveReviewOwnerAgentId ( evidence . sourceIssue , evidence . sourceAgent ) ;
let review : Awaited < ReturnType < typeof issuesSvc.create > > ;
try {
review = await issuesSvc . create ( evidence . sourceIssue . companyId , {
title : ` Review productivity for ${ evidence . sourceIssue . identifier ? ? evidence . sourceIssue . title } ` ,
description : buildReviewMarkdown ( evidence , opts . prefix ) ,
status : "todo" ,
priority : evidence.trigger === "long_active_duration" ? "medium" : "high" ,
parentId : evidence.sourceIssue.id ,
projectId : evidence.sourceIssue.projectId ,
goalId : evidence.sourceIssue.goalId ,
billingCode : evidence.sourceIssue.billingCode ,
assigneeAgentId : ownerAgentId ,
Guard cheap recovery model usage (#6371)
## Thinking Path
> - Paperclip is the control plane that coordinates AI-agent work
through issues, heartbeats, comments, approvals, and auditable recovery
paths.
> - The affected subsystem is heartbeat/recovery orchestration,
especially the optional cheap model profile used for operational
recovery overhead.
> - Cheap recovery should repair status and liveness, but it must not
become the worker lane that writes deliverables, continues source work,
or propagates cheap execution hints into downstream retries.
> - The gap was that cheap-profile hints could follow recovery wake
contexts and assignment overrides farther than intended, making real
work eligible to run on the cheap model.
> - This pull request separates status-only cheap recovery from normal
source-work continuations, adds route guards for deliverable mutations
during cheap status-only runs, and documents the invariant.
> - The benefit is safer retry/recovery behavior: cheap runs can clean
up control-plane state, while any remaining source work resumes through
a normal/original model path.
## What Changed
- Added recovery model-profile work classes so status-only recovery
carries explicit guard context and normal-model continuations scrub
cheap hints.
- Updated heartbeat, productivity review, liveness continuation, and
recovery service wakeups to request cheap only for bounded status-only
recovery work.
- Blocked cheap status-only recovery runs from writing issue documents,
plans, attachments, work products, or assigning downstream work back to
`modelProfile: "cheap"`.
- Added/updated server tests for cheap profile propagation,
artifact/document guards, route authorization, retry scheduling, and
successful-run handoff behavior.
- Documented the recovery model-profile lane in
`doc/SPEC-implementation.md` and `doc/execution-semantics.md`.
- After rebasing onto current `public-gh/master`, stabilized the new
`InstanceSidebar` plugin-filter tests so the PR check lane stays green.
## Verification
- Local: `pnpm exec vitest run --config vitest.config.ts
src/services/recovery/model-profile-hint.test.ts
src/__tests__/issue-agent-mutation-ownership-routes.test.ts
src/__tests__/issue-document-restore-routes.test.ts` from `server/` - 3
files, 37 tests passed after final edits.
- Local: `pnpm exec vitest run --config vitest.config.ts
src/__tests__/heartbeat-process-recovery.test.ts` from `server/` - 44
tests passed after rerunning the cleanup-sensitive file alone.
- Local: `pnpm --filter @paperclipai/ui exec vitest run
src/components/InstanceSidebar.test.tsx` - 4 tests passed.
- Local: `pnpm --filter @paperclipai/server typecheck` - passed.
- Local: `pnpm --filter @paperclipai/ui typecheck` - passed.
- PR checks on latest head `6f8c3b1380f5bd872c6f49f6f7188ecf3bb6d263` -
all green, including `verify`, build, typecheck,
server/general/serialized tests, e2e, Snyk, and policy.
- Greptile: pass 3 returned Confidence Score 5/5 with zero unresolved
Greptile review threads.
## Risks
- Medium risk: recovery behavior is intentionally stricter, so any path
that incorrectly relies on cheap recovery to keep doing source work will
now need to hand back to a normal-model run.
- Low migration risk: no schema changes.
- No product UI changes; the UI file touched is a test-only
stabilization after rebasing onto current `master`.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex coding agent, GPT-5 model family (`gpt-5`), tool use and
local code execution enabled; context window not exposed in this
environment.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots (N/A: no product UI changes)
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
2026-05-19 13:46:02 -05:00
assigneeAdapterOverrides : recoveryAssigneeAdapterOverrides ( "status_only" ) ,
[codex] Split backend control-plane QoL slice (#4700)
## Thinking Path
> - Paperclip is the control plane for autonomous AI companies, so
backend task ownership, recovery, review visibility, and company-scoped
limits need to stay enforceable without UI-only coupling.
> - Closed PR #4692 bundled those backend changes with UI workflow,
docs, skills, workflow, and lockfile churn.
> - PAP-2694 asks for a clean backend/control-plane slice from that
closed branch.
> - This branch starts from current `master` and mines only the `cli`,
`packages/db`, `packages/shared`, and `server` contracts/tests needed
for the backend behavior.
> - It explicitly excludes UI workflow/performance work,
`.github/workflows/pr.yml`, `pnpm-lock.yaml`, docs, skills,
package-script, adapter UI build-config, and perf fixture script
changes; the only UI files are fixture/test updates required by the
tightened shared `Company` contract.
> - The benefit is a smaller reviewable PR that preserves the
control-plane fixes while staying under Greptile s 100-file review
limit.
## What Changed
- Added company-scoped attachment-size limits through DB
schema/migrations, shared company portability contracts, CLI
import/export coverage, and server attachment upload enforcement.
- Added productivity review service/API behavior for no-comment streak,
long-active, and high-churn review issues, including request-depth
clamping and issue summary exposure.
- Hardened issue ownership and recovery/control-plane paths: peer-agent
mutation denial, issue tree pause/resume behavior, stranded recovery
origins, and related activity/test coverage.
- Preserved related backend contract updates for routine timestamp
variables and managed agent instruction bundles because they live in
shared/server contracts from the source branch.
- Addressed Greptile feedback by making `Company.attachmentMaxBytes`
non-optional, simplifying review request-depth clamping, fixing the
migration final newline, and enforcing the process-level attachment cap
as the final ceiling for uploads.
- Added minimal company fixtures needed for repo-wide typecheck/build
and kept the PR to 66 changed files with forbidden/non-slice paths
excluded.
## Verification
- `pnpm install --frozen-lockfile`
- `git diff --check origin/master..HEAD`
- `git diff --name-only origin/master..HEAD | wc -l` -> 66 files
- `git diff --name-only origin/master..HEAD -- .github/workflows/pr.yml
pnpm-lock.yaml package.json doc skills .agents scripts
packages/adapters` -> no output
- `pnpm exec vitest run --config vitest.config.ts
packages/shared/src/validators/issue.test.ts
packages/shared/src/routine-variables.test.ts
packages/shared/src/adapter-types.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
cli/src/__tests__/company.test.ts
server/src/__tests__/productivity-review-service.test.ts
server/src/__tests__/issue-tree-control-service.test.ts
server/src/__tests__/issue-tree-control-routes.test.ts
server/src/__tests__/issue-agent-mutation-ownership-routes.test.ts
server/src/__tests__/issue-attachment-routes.test.ts
server/src/__tests__/heartbeat-process-recovery.test.ts
server/src/__tests__/issues-service.test.ts` -> 12 files, 147 tests
passed
- `pnpm exec vitest run --config vitest.config.ts
cli/src/__tests__/company-delete.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
server/src/__tests__/productivity-review-service.test.ts` -> 3 files, 18
tests passed
- `pnpm exec vitest run --config vitest.config.ts
server/src/__tests__/issue-attachment-routes.test.ts` -> 1 file, 6 tests
passed
- `pnpm --filter @paperclipai/db typecheck && pnpm --filter
@paperclipai/shared typecheck && pnpm --filter @paperclipai/server
typecheck && pnpm --filter paperclipai typecheck`
- `pnpm --filter @paperclipai/server typecheck`
- `pnpm --filter @paperclipai/ui typecheck && pnpm --filter
@paperclipai/ui build`
## Risks
- Includes migrations `0073_shiny_salo.sql` and
`0074_striped_genesis.sql`; merge ordering matters if another PR adds
migrations first.
- This is intentionally backend-only apart from fixture/test updates
forced by shared type correctness; UI affordances from PR #4692 are not
present here and should land in separate UI slices.
- The worktree install emitted plugin SDK bin-link warnings for unbuilt
plugin packages, but the targeted tests and package typechecks completed
successfully.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected; check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex, GPT-5 coding agent, tool-enabled terminal/GitHub
workflow. Exact runtime context window was not exposed by the harness.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-04-28 16:46:45 -05:00
originKind : PRODUCTIVITY_REVIEW_ORIGIN_KIND ,
originId : evidence.sourceIssue.id ,
originFingerprint : productivityReviewFingerprint ( evidence . sourceIssue . id ) ,
requestDepth : clampIssueRequestDepth ( evidence . sourceIssue . requestDepth + 1 ) ,
} ) ;
} catch ( error ) {
const maybe = error as { code? : string ; constraint? : string ; message? : string } ;
const uniqueConflict = maybe . code === "23505" &&
(
maybe . constraint === "issues_active_productivity_review_uq" ||
typeof maybe . message === "string" && maybe . message . includes ( "issues_active_productivity_review_uq" )
) ;
if ( ! uniqueConflict ) throw error ;
const raced = await findOpenProductivityReview ( evidence . sourceIssue . companyId , evidence . sourceIssue . id ) ;
if ( ! raced ) throw error ;
return { kind : "existing" as const , reviewIssueId : raced.id } ;
}
[codex] Bound productivity review recovery loops (#4948)
## Thinking Path
> - Paperclip orchestrates AI agents for zero-human companies.
> - The heartbeat/productivity review subsystem detects when assigned
work is likely stuck or churning.
> - Productivity reviews are useful, but repeated reconciliation can
create noisy refresh comments or repeated review issues around the same
source issue.
> - That makes manager follow-up harder because the signal can get
buried under duplicate review activity.
> - This pull request bounds productivity review refreshes and creation
loops while preserving the existing escalation path.
> - The benefit is a quieter recovery loop that still surfaces stuck or
high-churn work for manager attention.
## What Changed
- Added refresh throttling for open productivity review issues,
including a one-hour default interval and a maximum of three refresh
comments per open review.
- Added a rolling 24-hour creation cap so completed/closed reviews
cannot immediately recreate review issues indefinitely for the same
source issue.
- Excluded cancelled productivity reviews from the creation cap so
manager cancellations do not silently suppress future legitimate
reviews.
- Preserved productivity review timestamps in deterministic test paths
and added targeted coverage for immediate refresh suppression, refresh
caps, creation caps, and cancelled-review exclusion.
## Verification
- `pnpm run preflight:workspace-links && pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- `pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- Greptile Review: 5/5 on commit
`bcf25832d0ffae25890b2ee7eed112d1c2d114fe` with review threads resolved.
- GitHub PR checks passed on the latest head: `policy`, `verify`, `e2e`,
`Greptile Review`, and `security/snyk (cryppadotta)`.
- Verified the branch is rebased onto `public-gh/master` with no
conflicts.
- Verified the diff does not include `pnpm-lock.yaml`, database schema
changes, or migrations.
## Risks
- Low-to-medium risk: this changes automation cadence for productivity
reviews. A truly stuck issue may receive fewer repeated refresh
comments, but the original review issue remains open and assigned for
manager action.
- No migration risk: this is server logic and tests only.
> Checked [`ROADMAP.md`](ROADMAP.md) for overlapping planned core work;
this is a targeted recovery-loop fix and does not add a new roadmap
feature.
## Model Used
- OpenAI Codex coding agent, GPT-5 model family, tool-using software
engineering mode. Exact context window is not exposed in this runtime.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots (not applicable; server-only change)
- [x] I have updated relevant documentation to reflect my changes (not
applicable; no user-facing docs or commands changed)
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-05-01 08:32:04 -05:00
await db
. update ( issues )
. set ( { createdAt : evidence.generatedAt , updatedAt : evidence.generatedAt } )
. where ( eq ( issues . id , review . id ) ) ;
[codex] Split backend control-plane QoL slice (#4700)
## Thinking Path
> - Paperclip is the control plane for autonomous AI companies, so
backend task ownership, recovery, review visibility, and company-scoped
limits need to stay enforceable without UI-only coupling.
> - Closed PR #4692 bundled those backend changes with UI workflow,
docs, skills, workflow, and lockfile churn.
> - PAP-2694 asks for a clean backend/control-plane slice from that
closed branch.
> - This branch starts from current `master` and mines only the `cli`,
`packages/db`, `packages/shared`, and `server` contracts/tests needed
for the backend behavior.
> - It explicitly excludes UI workflow/performance work,
`.github/workflows/pr.yml`, `pnpm-lock.yaml`, docs, skills,
package-script, adapter UI build-config, and perf fixture script
changes; the only UI files are fixture/test updates required by the
tightened shared `Company` contract.
> - The benefit is a smaller reviewable PR that preserves the
control-plane fixes while staying under Greptile s 100-file review
limit.
## What Changed
- Added company-scoped attachment-size limits through DB
schema/migrations, shared company portability contracts, CLI
import/export coverage, and server attachment upload enforcement.
- Added productivity review service/API behavior for no-comment streak,
long-active, and high-churn review issues, including request-depth
clamping and issue summary exposure.
- Hardened issue ownership and recovery/control-plane paths: peer-agent
mutation denial, issue tree pause/resume behavior, stranded recovery
origins, and related activity/test coverage.
- Preserved related backend contract updates for routine timestamp
variables and managed agent instruction bundles because they live in
shared/server contracts from the source branch.
- Addressed Greptile feedback by making `Company.attachmentMaxBytes`
non-optional, simplifying review request-depth clamping, fixing the
migration final newline, and enforcing the process-level attachment cap
as the final ceiling for uploads.
- Added minimal company fixtures needed for repo-wide typecheck/build
and kept the PR to 66 changed files with forbidden/non-slice paths
excluded.
## Verification
- `pnpm install --frozen-lockfile`
- `git diff --check origin/master..HEAD`
- `git diff --name-only origin/master..HEAD | wc -l` -> 66 files
- `git diff --name-only origin/master..HEAD -- .github/workflows/pr.yml
pnpm-lock.yaml package.json doc skills .agents scripts
packages/adapters` -> no output
- `pnpm exec vitest run --config vitest.config.ts
packages/shared/src/validators/issue.test.ts
packages/shared/src/routine-variables.test.ts
packages/shared/src/adapter-types.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
cli/src/__tests__/company.test.ts
server/src/__tests__/productivity-review-service.test.ts
server/src/__tests__/issue-tree-control-service.test.ts
server/src/__tests__/issue-tree-control-routes.test.ts
server/src/__tests__/issue-agent-mutation-ownership-routes.test.ts
server/src/__tests__/issue-attachment-routes.test.ts
server/src/__tests__/heartbeat-process-recovery.test.ts
server/src/__tests__/issues-service.test.ts` -> 12 files, 147 tests
passed
- `pnpm exec vitest run --config vitest.config.ts
cli/src/__tests__/company-delete.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
server/src/__tests__/productivity-review-service.test.ts` -> 3 files, 18
tests passed
- `pnpm exec vitest run --config vitest.config.ts
server/src/__tests__/issue-attachment-routes.test.ts` -> 1 file, 6 tests
passed
- `pnpm --filter @paperclipai/db typecheck && pnpm --filter
@paperclipai/shared typecheck && pnpm --filter @paperclipai/server
typecheck && pnpm --filter paperclipai typecheck`
- `pnpm --filter @paperclipai/server typecheck`
- `pnpm --filter @paperclipai/ui typecheck && pnpm --filter
@paperclipai/ui build`
## Risks
- Includes migrations `0073_shiny_salo.sql` and
`0074_striped_genesis.sql`; merge ordering matters if another PR adds
migrations first.
- This is intentionally backend-only apart from fixture/test updates
forced by shared type correctness; UI affordances from PR #4692 are not
present here and should land in separate UI slices.
- The worktree install emitted plugin SDK bin-link warnings for unbuilt
plugin packages, but the targeted tests and package typechecks completed
successfully.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected; check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex, GPT-5 coding agent, tool-enabled terminal/GitHub
workflow. Exact runtime context window was not exposed by the harness.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-04-28 16:46:45 -05:00
await logActivity ( db , {
companyId : evidence.sourceIssue.companyId ,
actorType : "system" ,
actorId : "system" ,
action : "issue.productivity_review_created" ,
entityType : "issue" ,
entityId : review.id ,
agentId : ownerAgentId ,
details : {
source : "productivity_review.reconcile" ,
sourceIssueId : evidence.sourceIssue.id ,
trigger : evidence.trigger ,
noCommentStreak : evidence.noCommentStreak ,
runCountLastHour : evidence.runCountLastHour ,
commentCountLastHour : evidence.commentCountLastHour ,
} ,
} ) ;
if ( ownerAgentId && deps ? . enqueueWakeup ) {
await deps . enqueueWakeup ( ownerAgentId , {
source : "assignment" ,
triggerDetail : "system" ,
reason : "issue_assigned" ,
Add recovery handoff system notices (#5289)
## Thinking Path
> - Paperclip orchestrates AI agents for zero-human companies.
> - Agent runs can end productively while the source issue still lacks a
durable final disposition.
> - That leaves the control plane unsure whether to resume, escalate, or
close the work.
> - Issue comments also need a presentation contract so system-authored
recovery notices can render as first-class thread messages without
overloading normal comments.
> - This pull request adds successful-run handoff recovery, comment
presentation metadata, and system notice rendering.
> - The benefit is stricter task liveness with clearer operator-facing
recovery state.
## What Changed
- Added successful-run handoff decisions, wake payloads, escalation
behavior, and recovery tests.
- Added issue comment presentation metadata with migration
`0078_white_darwin.sql` and shared/server/company portability support.
- Rendered recovery/system notices in issue chat with dedicated UI
components, fixtures, tests, and storybook/lab coverage.
- Included the current recovery model-profile hint patch so automatic
recovery follow-ups use the cheap profile.
## Verification
- `pnpm install --frozen-lockfile`
- `pnpm exec vitest run
server/src/services/recovery/successful-run-handoff.test.ts
ui/src/components/SystemNotice.test.tsx
ui/src/lib/system-notice-comment.test.ts
ui/src/components/IssueChatThreadSystemNotice.test.tsx`
## Risks
- Migration-bearing PR: merge this before any other branch that might
later add a migration.
- The branch touches both recovery services and issue-thread rendering,
so review should pay attention to recovery wake idempotency and comment
metadata compatibility.
## Model Used
- OpenAI GPT-5 Codex via Paperclip `codex_local` adapter, with
shell/git/GitHub CLI tool use.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-05-06 06:05:58 -05:00
payload : withRecoveryModelProfileHint ( {
[codex] Split backend control-plane QoL slice (#4700)
## Thinking Path
> - Paperclip is the control plane for autonomous AI companies, so
backend task ownership, recovery, review visibility, and company-scoped
limits need to stay enforceable without UI-only coupling.
> - Closed PR #4692 bundled those backend changes with UI workflow,
docs, skills, workflow, and lockfile churn.
> - PAP-2694 asks for a clean backend/control-plane slice from that
closed branch.
> - This branch starts from current `master` and mines only the `cli`,
`packages/db`, `packages/shared`, and `server` contracts/tests needed
for the backend behavior.
> - It explicitly excludes UI workflow/performance work,
`.github/workflows/pr.yml`, `pnpm-lock.yaml`, docs, skills,
package-script, adapter UI build-config, and perf fixture script
changes; the only UI files are fixture/test updates required by the
tightened shared `Company` contract.
> - The benefit is a smaller reviewable PR that preserves the
control-plane fixes while staying under Greptile s 100-file review
limit.
## What Changed
- Added company-scoped attachment-size limits through DB
schema/migrations, shared company portability contracts, CLI
import/export coverage, and server attachment upload enforcement.
- Added productivity review service/API behavior for no-comment streak,
long-active, and high-churn review issues, including request-depth
clamping and issue summary exposure.
- Hardened issue ownership and recovery/control-plane paths: peer-agent
mutation denial, issue tree pause/resume behavior, stranded recovery
origins, and related activity/test coverage.
- Preserved related backend contract updates for routine timestamp
variables and managed agent instruction bundles because they live in
shared/server contracts from the source branch.
- Addressed Greptile feedback by making `Company.attachmentMaxBytes`
non-optional, simplifying review request-depth clamping, fixing the
migration final newline, and enforcing the process-level attachment cap
as the final ceiling for uploads.
- Added minimal company fixtures needed for repo-wide typecheck/build
and kept the PR to 66 changed files with forbidden/non-slice paths
excluded.
## Verification
- `pnpm install --frozen-lockfile`
- `git diff --check origin/master..HEAD`
- `git diff --name-only origin/master..HEAD | wc -l` -> 66 files
- `git diff --name-only origin/master..HEAD -- .github/workflows/pr.yml
pnpm-lock.yaml package.json doc skills .agents scripts
packages/adapters` -> no output
- `pnpm exec vitest run --config vitest.config.ts
packages/shared/src/validators/issue.test.ts
packages/shared/src/routine-variables.test.ts
packages/shared/src/adapter-types.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
cli/src/__tests__/company.test.ts
server/src/__tests__/productivity-review-service.test.ts
server/src/__tests__/issue-tree-control-service.test.ts
server/src/__tests__/issue-tree-control-routes.test.ts
server/src/__tests__/issue-agent-mutation-ownership-routes.test.ts
server/src/__tests__/issue-attachment-routes.test.ts
server/src/__tests__/heartbeat-process-recovery.test.ts
server/src/__tests__/issues-service.test.ts` -> 12 files, 147 tests
passed
- `pnpm exec vitest run --config vitest.config.ts
cli/src/__tests__/company-delete.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
server/src/__tests__/productivity-review-service.test.ts` -> 3 files, 18
tests passed
- `pnpm exec vitest run --config vitest.config.ts
server/src/__tests__/issue-attachment-routes.test.ts` -> 1 file, 6 tests
passed
- `pnpm --filter @paperclipai/db typecheck && pnpm --filter
@paperclipai/shared typecheck && pnpm --filter @paperclipai/server
typecheck && pnpm --filter paperclipai typecheck`
- `pnpm --filter @paperclipai/server typecheck`
- `pnpm --filter @paperclipai/ui typecheck && pnpm --filter
@paperclipai/ui build`
## Risks
- Includes migrations `0073_shiny_salo.sql` and
`0074_striped_genesis.sql`; merge ordering matters if another PR adds
migrations first.
- This is intentionally backend-only apart from fixture/test updates
forced by shared type correctness; UI affordances from PR #4692 are not
present here and should land in separate UI slices.
- The worktree install emitted plugin SDK bin-link warnings for unbuilt
plugin packages, but the targeted tests and package typechecks completed
successfully.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected; check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex, GPT-5 coding agent, tool-enabled terminal/GitHub
workflow. Exact runtime context window was not exposed by the harness.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-04-28 16:46:45 -05:00
issueId : review.id ,
sourceIssueId : evidence.sourceIssue.id ,
trigger : evidence.trigger ,
Guard cheap recovery model usage (#6371)
## Thinking Path
> - Paperclip is the control plane that coordinates AI-agent work
through issues, heartbeats, comments, approvals, and auditable recovery
paths.
> - The affected subsystem is heartbeat/recovery orchestration,
especially the optional cheap model profile used for operational
recovery overhead.
> - Cheap recovery should repair status and liveness, but it must not
become the worker lane that writes deliverables, continues source work,
or propagates cheap execution hints into downstream retries.
> - The gap was that cheap-profile hints could follow recovery wake
contexts and assignment overrides farther than intended, making real
work eligible to run on the cheap model.
> - This pull request separates status-only cheap recovery from normal
source-work continuations, adds route guards for deliverable mutations
during cheap status-only runs, and documents the invariant.
> - The benefit is safer retry/recovery behavior: cheap runs can clean
up control-plane state, while any remaining source work resumes through
a normal/original model path.
## What Changed
- Added recovery model-profile work classes so status-only recovery
carries explicit guard context and normal-model continuations scrub
cheap hints.
- Updated heartbeat, productivity review, liveness continuation, and
recovery service wakeups to request cheap only for bounded status-only
recovery work.
- Blocked cheap status-only recovery runs from writing issue documents,
plans, attachments, work products, or assigning downstream work back to
`modelProfile: "cheap"`.
- Added/updated server tests for cheap profile propagation,
artifact/document guards, route authorization, retry scheduling, and
successful-run handoff behavior.
- Documented the recovery model-profile lane in
`doc/SPEC-implementation.md` and `doc/execution-semantics.md`.
- After rebasing onto current `public-gh/master`, stabilized the new
`InstanceSidebar` plugin-filter tests so the PR check lane stays green.
## Verification
- Local: `pnpm exec vitest run --config vitest.config.ts
src/services/recovery/model-profile-hint.test.ts
src/__tests__/issue-agent-mutation-ownership-routes.test.ts
src/__tests__/issue-document-restore-routes.test.ts` from `server/` - 3
files, 37 tests passed after final edits.
- Local: `pnpm exec vitest run --config vitest.config.ts
src/__tests__/heartbeat-process-recovery.test.ts` from `server/` - 44
tests passed after rerunning the cleanup-sensitive file alone.
- Local: `pnpm --filter @paperclipai/ui exec vitest run
src/components/InstanceSidebar.test.tsx` - 4 tests passed.
- Local: `pnpm --filter @paperclipai/server typecheck` - passed.
- Local: `pnpm --filter @paperclipai/ui typecheck` - passed.
- PR checks on latest head `6f8c3b1380f5bd872c6f49f6f7188ecf3bb6d263` -
all green, including `verify`, build, typecheck,
server/general/serialized tests, e2e, Snyk, and policy.
- Greptile: pass 3 returned Confidence Score 5/5 with zero unresolved
Greptile review threads.
## Risks
- Medium risk: recovery behavior is intentionally stricter, so any path
that incorrectly relies on cheap recovery to keep doing source work will
now need to hand back to a normal-model run.
- Low migration risk: no schema changes.
- No product UI changes; the UI file touched is a test-only
stabilization after rebasing onto current `master`.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex coding agent, GPT-5 model family (`gpt-5`), tool use and
local code execution enabled; context window not exposed in this
environment.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots (N/A: no product UI changes)
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
2026-05-19 13:46:02 -05:00
} , "status_only" ) ,
[codex] Split backend control-plane QoL slice (#4700)
## Thinking Path
> - Paperclip is the control plane for autonomous AI companies, so
backend task ownership, recovery, review visibility, and company-scoped
limits need to stay enforceable without UI-only coupling.
> - Closed PR #4692 bundled those backend changes with UI workflow,
docs, skills, workflow, and lockfile churn.
> - PAP-2694 asks for a clean backend/control-plane slice from that
closed branch.
> - This branch starts from current `master` and mines only the `cli`,
`packages/db`, `packages/shared`, and `server` contracts/tests needed
for the backend behavior.
> - It explicitly excludes UI workflow/performance work,
`.github/workflows/pr.yml`, `pnpm-lock.yaml`, docs, skills,
package-script, adapter UI build-config, and perf fixture script
changes; the only UI files are fixture/test updates required by the
tightened shared `Company` contract.
> - The benefit is a smaller reviewable PR that preserves the
control-plane fixes while staying under Greptile s 100-file review
limit.
## What Changed
- Added company-scoped attachment-size limits through DB
schema/migrations, shared company portability contracts, CLI
import/export coverage, and server attachment upload enforcement.
- Added productivity review service/API behavior for no-comment streak,
long-active, and high-churn review issues, including request-depth
clamping and issue summary exposure.
- Hardened issue ownership and recovery/control-plane paths: peer-agent
mutation denial, issue tree pause/resume behavior, stranded recovery
origins, and related activity/test coverage.
- Preserved related backend contract updates for routine timestamp
variables and managed agent instruction bundles because they live in
shared/server contracts from the source branch.
- Addressed Greptile feedback by making `Company.attachmentMaxBytes`
non-optional, simplifying review request-depth clamping, fixing the
migration final newline, and enforcing the process-level attachment cap
as the final ceiling for uploads.
- Added minimal company fixtures needed for repo-wide typecheck/build
and kept the PR to 66 changed files with forbidden/non-slice paths
excluded.
## Verification
- `pnpm install --frozen-lockfile`
- `git diff --check origin/master..HEAD`
- `git diff --name-only origin/master..HEAD | wc -l` -> 66 files
- `git diff --name-only origin/master..HEAD -- .github/workflows/pr.yml
pnpm-lock.yaml package.json doc skills .agents scripts
packages/adapters` -> no output
- `pnpm exec vitest run --config vitest.config.ts
packages/shared/src/validators/issue.test.ts
packages/shared/src/routine-variables.test.ts
packages/shared/src/adapter-types.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
cli/src/__tests__/company.test.ts
server/src/__tests__/productivity-review-service.test.ts
server/src/__tests__/issue-tree-control-service.test.ts
server/src/__tests__/issue-tree-control-routes.test.ts
server/src/__tests__/issue-agent-mutation-ownership-routes.test.ts
server/src/__tests__/issue-attachment-routes.test.ts
server/src/__tests__/heartbeat-process-recovery.test.ts
server/src/__tests__/issues-service.test.ts` -> 12 files, 147 tests
passed
- `pnpm exec vitest run --config vitest.config.ts
cli/src/__tests__/company-delete.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
server/src/__tests__/productivity-review-service.test.ts` -> 3 files, 18
tests passed
- `pnpm exec vitest run --config vitest.config.ts
server/src/__tests__/issue-attachment-routes.test.ts` -> 1 file, 6 tests
passed
- `pnpm --filter @paperclipai/db typecheck && pnpm --filter
@paperclipai/shared typecheck && pnpm --filter @paperclipai/server
typecheck && pnpm --filter paperclipai typecheck`
- `pnpm --filter @paperclipai/server typecheck`
- `pnpm --filter @paperclipai/ui typecheck && pnpm --filter
@paperclipai/ui build`
## Risks
- Includes migrations `0073_shiny_salo.sql` and
`0074_striped_genesis.sql`; merge ordering matters if another PR adds
migrations first.
- This is intentionally backend-only apart from fixture/test updates
forced by shared type correctness; UI affordances from PR #4692 are not
present here and should land in separate UI slices.
- The worktree install emitted plugin SDK bin-link warnings for unbuilt
plugin packages, but the targeted tests and package typechecks completed
successfully.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected; check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex, GPT-5 coding agent, tool-enabled terminal/GitHub
workflow. Exact runtime context window was not exposed by the harness.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-04-28 16:46:45 -05:00
requestedByActorType : "system" ,
requestedByActorId : "productivity_review" ,
Add recovery handoff system notices (#5289)
## Thinking Path
> - Paperclip orchestrates AI agents for zero-human companies.
> - Agent runs can end productively while the source issue still lacks a
durable final disposition.
> - That leaves the control plane unsure whether to resume, escalate, or
close the work.
> - Issue comments also need a presentation contract so system-authored
recovery notices can render as first-class thread messages without
overloading normal comments.
> - This pull request adds successful-run handoff recovery, comment
presentation metadata, and system notice rendering.
> - The benefit is stricter task liveness with clearer operator-facing
recovery state.
## What Changed
- Added successful-run handoff decisions, wake payloads, escalation
behavior, and recovery tests.
- Added issue comment presentation metadata with migration
`0078_white_darwin.sql` and shared/server/company portability support.
- Rendered recovery/system notices in issue chat with dedicated UI
components, fixtures, tests, and storybook/lab coverage.
- Included the current recovery model-profile hint patch so automatic
recovery follow-ups use the cheap profile.
## Verification
- `pnpm install --frozen-lockfile`
- `pnpm exec vitest run
server/src/services/recovery/successful-run-handoff.test.ts
ui/src/components/SystemNotice.test.tsx
ui/src/lib/system-notice-comment.test.ts
ui/src/components/IssueChatThreadSystemNotice.test.tsx`
## Risks
- Migration-bearing PR: merge this before any other branch that might
later add a migration.
- The branch touches both recovery services and issue-thread rendering,
so review should pay attention to recovery wake idempotency and comment
metadata compatibility.
## Model Used
- OpenAI GPT-5 Codex via Paperclip `codex_local` adapter, with
shell/git/GitHub CLI tool use.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-05-06 06:05:58 -05:00
contextSnapshot : withRecoveryModelProfileHint ( {
[codex] Split backend control-plane QoL slice (#4700)
## Thinking Path
> - Paperclip is the control plane for autonomous AI companies, so
backend task ownership, recovery, review visibility, and company-scoped
limits need to stay enforceable without UI-only coupling.
> - Closed PR #4692 bundled those backend changes with UI workflow,
docs, skills, workflow, and lockfile churn.
> - PAP-2694 asks for a clean backend/control-plane slice from that
closed branch.
> - This branch starts from current `master` and mines only the `cli`,
`packages/db`, `packages/shared`, and `server` contracts/tests needed
for the backend behavior.
> - It explicitly excludes UI workflow/performance work,
`.github/workflows/pr.yml`, `pnpm-lock.yaml`, docs, skills,
package-script, adapter UI build-config, and perf fixture script
changes; the only UI files are fixture/test updates required by the
tightened shared `Company` contract.
> - The benefit is a smaller reviewable PR that preserves the
control-plane fixes while staying under Greptile s 100-file review
limit.
## What Changed
- Added company-scoped attachment-size limits through DB
schema/migrations, shared company portability contracts, CLI
import/export coverage, and server attachment upload enforcement.
- Added productivity review service/API behavior for no-comment streak,
long-active, and high-churn review issues, including request-depth
clamping and issue summary exposure.
- Hardened issue ownership and recovery/control-plane paths: peer-agent
mutation denial, issue tree pause/resume behavior, stranded recovery
origins, and related activity/test coverage.
- Preserved related backend contract updates for routine timestamp
variables and managed agent instruction bundles because they live in
shared/server contracts from the source branch.
- Addressed Greptile feedback by making `Company.attachmentMaxBytes`
non-optional, simplifying review request-depth clamping, fixing the
migration final newline, and enforcing the process-level attachment cap
as the final ceiling for uploads.
- Added minimal company fixtures needed for repo-wide typecheck/build
and kept the PR to 66 changed files with forbidden/non-slice paths
excluded.
## Verification
- `pnpm install --frozen-lockfile`
- `git diff --check origin/master..HEAD`
- `git diff --name-only origin/master..HEAD | wc -l` -> 66 files
- `git diff --name-only origin/master..HEAD -- .github/workflows/pr.yml
pnpm-lock.yaml package.json doc skills .agents scripts
packages/adapters` -> no output
- `pnpm exec vitest run --config vitest.config.ts
packages/shared/src/validators/issue.test.ts
packages/shared/src/routine-variables.test.ts
packages/shared/src/adapter-types.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
cli/src/__tests__/company.test.ts
server/src/__tests__/productivity-review-service.test.ts
server/src/__tests__/issue-tree-control-service.test.ts
server/src/__tests__/issue-tree-control-routes.test.ts
server/src/__tests__/issue-agent-mutation-ownership-routes.test.ts
server/src/__tests__/issue-attachment-routes.test.ts
server/src/__tests__/heartbeat-process-recovery.test.ts
server/src/__tests__/issues-service.test.ts` -> 12 files, 147 tests
passed
- `pnpm exec vitest run --config vitest.config.ts
cli/src/__tests__/company-delete.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
server/src/__tests__/productivity-review-service.test.ts` -> 3 files, 18
tests passed
- `pnpm exec vitest run --config vitest.config.ts
server/src/__tests__/issue-attachment-routes.test.ts` -> 1 file, 6 tests
passed
- `pnpm --filter @paperclipai/db typecheck && pnpm --filter
@paperclipai/shared typecheck && pnpm --filter @paperclipai/server
typecheck && pnpm --filter paperclipai typecheck`
- `pnpm --filter @paperclipai/server typecheck`
- `pnpm --filter @paperclipai/ui typecheck && pnpm --filter
@paperclipai/ui build`
## Risks
- Includes migrations `0073_shiny_salo.sql` and
`0074_striped_genesis.sql`; merge ordering matters if another PR adds
migrations first.
- This is intentionally backend-only apart from fixture/test updates
forced by shared type correctness; UI affordances from PR #4692 are not
present here and should land in separate UI slices.
- The worktree install emitted plugin SDK bin-link warnings for unbuilt
plugin packages, but the targeted tests and package typechecks completed
successfully.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected; check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex, GPT-5 coding agent, tool-enabled terminal/GitHub
workflow. Exact runtime context window was not exposed by the harness.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-04-28 16:46:45 -05:00
issueId : review.id ,
taskId : review.id ,
wakeReason : "issue_assigned" ,
source : PRODUCTIVITY_REVIEW_ORIGIN_KIND ,
sourceIssueId : evidence.sourceIssue.id ,
productivityReviewTrigger : evidence.trigger ,
Guard cheap recovery model usage (#6371)
## Thinking Path
> - Paperclip is the control plane that coordinates AI-agent work
through issues, heartbeats, comments, approvals, and auditable recovery
paths.
> - The affected subsystem is heartbeat/recovery orchestration,
especially the optional cheap model profile used for operational
recovery overhead.
> - Cheap recovery should repair status and liveness, but it must not
become the worker lane that writes deliverables, continues source work,
or propagates cheap execution hints into downstream retries.
> - The gap was that cheap-profile hints could follow recovery wake
contexts and assignment overrides farther than intended, making real
work eligible to run on the cheap model.
> - This pull request separates status-only cheap recovery from normal
source-work continuations, adds route guards for deliverable mutations
during cheap status-only runs, and documents the invariant.
> - The benefit is safer retry/recovery behavior: cheap runs can clean
up control-plane state, while any remaining source work resumes through
a normal/original model path.
## What Changed
- Added recovery model-profile work classes so status-only recovery
carries explicit guard context and normal-model continuations scrub
cheap hints.
- Updated heartbeat, productivity review, liveness continuation, and
recovery service wakeups to request cheap only for bounded status-only
recovery work.
- Blocked cheap status-only recovery runs from writing issue documents,
plans, attachments, work products, or assigning downstream work back to
`modelProfile: "cheap"`.
- Added/updated server tests for cheap profile propagation,
artifact/document guards, route authorization, retry scheduling, and
successful-run handoff behavior.
- Documented the recovery model-profile lane in
`doc/SPEC-implementation.md` and `doc/execution-semantics.md`.
- After rebasing onto current `public-gh/master`, stabilized the new
`InstanceSidebar` plugin-filter tests so the PR check lane stays green.
## Verification
- Local: `pnpm exec vitest run --config vitest.config.ts
src/services/recovery/model-profile-hint.test.ts
src/__tests__/issue-agent-mutation-ownership-routes.test.ts
src/__tests__/issue-document-restore-routes.test.ts` from `server/` - 3
files, 37 tests passed after final edits.
- Local: `pnpm exec vitest run --config vitest.config.ts
src/__tests__/heartbeat-process-recovery.test.ts` from `server/` - 44
tests passed after rerunning the cleanup-sensitive file alone.
- Local: `pnpm --filter @paperclipai/ui exec vitest run
src/components/InstanceSidebar.test.tsx` - 4 tests passed.
- Local: `pnpm --filter @paperclipai/server typecheck` - passed.
- Local: `pnpm --filter @paperclipai/ui typecheck` - passed.
- PR checks on latest head `6f8c3b1380f5bd872c6f49f6f7188ecf3bb6d263` -
all green, including `verify`, build, typecheck,
server/general/serialized tests, e2e, Snyk, and policy.
- Greptile: pass 3 returned Confidence Score 5/5 with zero unresolved
Greptile review threads.
## Risks
- Medium risk: recovery behavior is intentionally stricter, so any path
that incorrectly relies on cheap recovery to keep doing source work will
now need to hand back to a normal-model run.
- Low migration risk: no schema changes.
- No product UI changes; the UI file touched is a test-only
stabilization after rebasing onto current `master`.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex coding agent, GPT-5 model family (`gpt-5`), tool use and
local code execution enabled; context window not exposed in this
environment.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots (N/A: no product UI changes)
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
2026-05-19 13:46:02 -05:00
} , "status_only" ) ,
[codex] Split backend control-plane QoL slice (#4700)
## Thinking Path
> - Paperclip is the control plane for autonomous AI companies, so
backend task ownership, recovery, review visibility, and company-scoped
limits need to stay enforceable without UI-only coupling.
> - Closed PR #4692 bundled those backend changes with UI workflow,
docs, skills, workflow, and lockfile churn.
> - PAP-2694 asks for a clean backend/control-plane slice from that
closed branch.
> - This branch starts from current `master` and mines only the `cli`,
`packages/db`, `packages/shared`, and `server` contracts/tests needed
for the backend behavior.
> - It explicitly excludes UI workflow/performance work,
`.github/workflows/pr.yml`, `pnpm-lock.yaml`, docs, skills,
package-script, adapter UI build-config, and perf fixture script
changes; the only UI files are fixture/test updates required by the
tightened shared `Company` contract.
> - The benefit is a smaller reviewable PR that preserves the
control-plane fixes while staying under Greptile s 100-file review
limit.
## What Changed
- Added company-scoped attachment-size limits through DB
schema/migrations, shared company portability contracts, CLI
import/export coverage, and server attachment upload enforcement.
- Added productivity review service/API behavior for no-comment streak,
long-active, and high-churn review issues, including request-depth
clamping and issue summary exposure.
- Hardened issue ownership and recovery/control-plane paths: peer-agent
mutation denial, issue tree pause/resume behavior, stranded recovery
origins, and related activity/test coverage.
- Preserved related backend contract updates for routine timestamp
variables and managed agent instruction bundles because they live in
shared/server contracts from the source branch.
- Addressed Greptile feedback by making `Company.attachmentMaxBytes`
non-optional, simplifying review request-depth clamping, fixing the
migration final newline, and enforcing the process-level attachment cap
as the final ceiling for uploads.
- Added minimal company fixtures needed for repo-wide typecheck/build
and kept the PR to 66 changed files with forbidden/non-slice paths
excluded.
## Verification
- `pnpm install --frozen-lockfile`
- `git diff --check origin/master..HEAD`
- `git diff --name-only origin/master..HEAD | wc -l` -> 66 files
- `git diff --name-only origin/master..HEAD -- .github/workflows/pr.yml
pnpm-lock.yaml package.json doc skills .agents scripts
packages/adapters` -> no output
- `pnpm exec vitest run --config vitest.config.ts
packages/shared/src/validators/issue.test.ts
packages/shared/src/routine-variables.test.ts
packages/shared/src/adapter-types.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
cli/src/__tests__/company.test.ts
server/src/__tests__/productivity-review-service.test.ts
server/src/__tests__/issue-tree-control-service.test.ts
server/src/__tests__/issue-tree-control-routes.test.ts
server/src/__tests__/issue-agent-mutation-ownership-routes.test.ts
server/src/__tests__/issue-attachment-routes.test.ts
server/src/__tests__/heartbeat-process-recovery.test.ts
server/src/__tests__/issues-service.test.ts` -> 12 files, 147 tests
passed
- `pnpm exec vitest run --config vitest.config.ts
cli/src/__tests__/company-delete.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
server/src/__tests__/productivity-review-service.test.ts` -> 3 files, 18
tests passed
- `pnpm exec vitest run --config vitest.config.ts
server/src/__tests__/issue-attachment-routes.test.ts` -> 1 file, 6 tests
passed
- `pnpm --filter @paperclipai/db typecheck && pnpm --filter
@paperclipai/shared typecheck && pnpm --filter @paperclipai/server
typecheck && pnpm --filter paperclipai typecheck`
- `pnpm --filter @paperclipai/server typecheck`
- `pnpm --filter @paperclipai/ui typecheck && pnpm --filter
@paperclipai/ui build`
## Risks
- Includes migrations `0073_shiny_salo.sql` and
`0074_striped_genesis.sql`; merge ordering matters if another PR adds
migrations first.
- This is intentionally backend-only apart from fixture/test updates
forced by shared type correctness; UI affordances from PR #4692 are not
present here and should land in separate UI slices.
- The worktree install emitted plugin SDK bin-link warnings for unbuilt
plugin packages, but the targeted tests and package typechecks completed
successfully.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected; check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex, GPT-5 coding agent, tool-enabled terminal/GitHub
workflow. Exact runtime context window was not exposed by the harness.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-04-28 16:46:45 -05:00
} ) ;
}
return { kind : "created" as const , reviewIssueId : review.id } ;
}
async function reconcileProductivityReviews ( opts ? : {
now? : Date ;
companyId? : string ;
thresholds? : Partial < ProductivityReviewThresholds > ;
} ) {
const now = opts ? . now ? ? new Date ( ) ;
const thresholds = buildThresholds ( opts ? . thresholds ) ;
const candidates = await db
. select ( )
. from ( issues )
. where (
and (
opts ? . companyId ? eq ( issues . companyId , opts . companyId ) : undefined ,
isNull ( issues . hiddenAt ) ,
isNull ( issues . assigneeUserId ) ,
inArray ( issues . status , [ "todo" , "in_progress" ] ) ,
sql ` ${ issues . assigneeAgentId } is not null ` ,
sql ` ${ issues . originKind } <> ${ PRODUCTIVITY_REVIEW_ORIGIN_KIND } ` ,
) ,
)
. orderBy ( asc ( issues . updatedAt ) , asc ( issues . id ) )
. limit ( MAX_CANDIDATE_ISSUES ) ;
const result = {
scanned : candidates.length ,
created : 0 ,
updated : 0 ,
existing : 0 ,
snoozed : 0 ,
[codex] Bound productivity review recovery loops (#4948)
## Thinking Path
> - Paperclip orchestrates AI agents for zero-human companies.
> - The heartbeat/productivity review subsystem detects when assigned
work is likely stuck or churning.
> - Productivity reviews are useful, but repeated reconciliation can
create noisy refresh comments or repeated review issues around the same
source issue.
> - That makes manager follow-up harder because the signal can get
buried under duplicate review activity.
> - This pull request bounds productivity review refreshes and creation
loops while preserving the existing escalation path.
> - The benefit is a quieter recovery loop that still surfaces stuck or
high-churn work for manager attention.
## What Changed
- Added refresh throttling for open productivity review issues,
including a one-hour default interval and a maximum of three refresh
comments per open review.
- Added a rolling 24-hour creation cap so completed/closed reviews
cannot immediately recreate review issues indefinitely for the same
source issue.
- Excluded cancelled productivity reviews from the creation cap so
manager cancellations do not silently suppress future legitimate
reviews.
- Preserved productivity review timestamps in deterministic test paths
and added targeted coverage for immediate refresh suppression, refresh
caps, creation caps, and cancelled-review exclusion.
## Verification
- `pnpm run preflight:workspace-links && pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- `pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- Greptile Review: 5/5 on commit
`bcf25832d0ffae25890b2ee7eed112d1c2d114fe` with review threads resolved.
- GitHub PR checks passed on the latest head: `policy`, `verify`, `e2e`,
`Greptile Review`, and `security/snyk (cryppadotta)`.
- Verified the branch is rebased onto `public-gh/master` with no
conflicts.
- Verified the diff does not include `pnpm-lock.yaml`, database schema
changes, or migrations.
## Risks
- Low-to-medium risk: this changes automation cadence for productivity
reviews. A truly stuck issue may receive fewer repeated refresh
comments, but the original review issue remains open and assigned for
manager action.
- No migration risk: this is server logic and tests only.
> Checked [`ROADMAP.md`](ROADMAP.md) for overlapping planned core work;
this is a targeted recovery-loop fix and does not add a new roadmap
feature.
## Model Used
- OpenAI Codex coding agent, GPT-5 model family, tool-using software
engineering mode. Exact context window is not exposed in this runtime.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots (not applicable; server-only change)
- [x] I have updated relevant documentation to reflect my changes (not
applicable; no user-facing docs or commands changed)
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-05-01 08:32:04 -05:00
creationCapped : 0 ,
[codex] Split backend control-plane QoL slice (#4700)
## Thinking Path
> - Paperclip is the control plane for autonomous AI companies, so
backend task ownership, recovery, review visibility, and company-scoped
limits need to stay enforceable without UI-only coupling.
> - Closed PR #4692 bundled those backend changes with UI workflow,
docs, skills, workflow, and lockfile churn.
> - PAP-2694 asks for a clean backend/control-plane slice from that
closed branch.
> - This branch starts from current `master` and mines only the `cli`,
`packages/db`, `packages/shared`, and `server` contracts/tests needed
for the backend behavior.
> - It explicitly excludes UI workflow/performance work,
`.github/workflows/pr.yml`, `pnpm-lock.yaml`, docs, skills,
package-script, adapter UI build-config, and perf fixture script
changes; the only UI files are fixture/test updates required by the
tightened shared `Company` contract.
> - The benefit is a smaller reviewable PR that preserves the
control-plane fixes while staying under Greptile s 100-file review
limit.
## What Changed
- Added company-scoped attachment-size limits through DB
schema/migrations, shared company portability contracts, CLI
import/export coverage, and server attachment upload enforcement.
- Added productivity review service/API behavior for no-comment streak,
long-active, and high-churn review issues, including request-depth
clamping and issue summary exposure.
- Hardened issue ownership and recovery/control-plane paths: peer-agent
mutation denial, issue tree pause/resume behavior, stranded recovery
origins, and related activity/test coverage.
- Preserved related backend contract updates for routine timestamp
variables and managed agent instruction bundles because they live in
shared/server contracts from the source branch.
- Addressed Greptile feedback by making `Company.attachmentMaxBytes`
non-optional, simplifying review request-depth clamping, fixing the
migration final newline, and enforcing the process-level attachment cap
as the final ceiling for uploads.
- Added minimal company fixtures needed for repo-wide typecheck/build
and kept the PR to 66 changed files with forbidden/non-slice paths
excluded.
## Verification
- `pnpm install --frozen-lockfile`
- `git diff --check origin/master..HEAD`
- `git diff --name-only origin/master..HEAD | wc -l` -> 66 files
- `git diff --name-only origin/master..HEAD -- .github/workflows/pr.yml
pnpm-lock.yaml package.json doc skills .agents scripts
packages/adapters` -> no output
- `pnpm exec vitest run --config vitest.config.ts
packages/shared/src/validators/issue.test.ts
packages/shared/src/routine-variables.test.ts
packages/shared/src/adapter-types.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
cli/src/__tests__/company.test.ts
server/src/__tests__/productivity-review-service.test.ts
server/src/__tests__/issue-tree-control-service.test.ts
server/src/__tests__/issue-tree-control-routes.test.ts
server/src/__tests__/issue-agent-mutation-ownership-routes.test.ts
server/src/__tests__/issue-attachment-routes.test.ts
server/src/__tests__/heartbeat-process-recovery.test.ts
server/src/__tests__/issues-service.test.ts` -> 12 files, 147 tests
passed
- `pnpm exec vitest run --config vitest.config.ts
cli/src/__tests__/company-delete.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
server/src/__tests__/productivity-review-service.test.ts` -> 3 files, 18
tests passed
- `pnpm exec vitest run --config vitest.config.ts
server/src/__tests__/issue-attachment-routes.test.ts` -> 1 file, 6 tests
passed
- `pnpm --filter @paperclipai/db typecheck && pnpm --filter
@paperclipai/shared typecheck && pnpm --filter @paperclipai/server
typecheck && pnpm --filter paperclipai typecheck`
- `pnpm --filter @paperclipai/server typecheck`
- `pnpm --filter @paperclipai/ui typecheck && pnpm --filter
@paperclipai/ui build`
## Risks
- Includes migrations `0073_shiny_salo.sql` and
`0074_striped_genesis.sql`; merge ordering matters if another PR adds
migrations first.
- This is intentionally backend-only apart from fixture/test updates
forced by shared type correctness; UI affordances from PR #4692 are not
present here and should land in separate UI slices.
- The worktree install emitted plugin SDK bin-link warnings for unbuilt
plugin packages, but the targeted tests and package typechecks completed
successfully.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected; check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex, GPT-5 coding agent, tool-enabled terminal/GitHub
workflow. Exact runtime context window was not exposed by the harness.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-04-28 16:46:45 -05:00
skipped : 0 ,
failed : 0 ,
reviewIssueIds : [ ] as string [ ] ,
failedIssueIds : [ ] as string [ ] ,
} ;
const prefixCache = new Map < string , string > ( ) ;
for ( const candidate of candidates ) {
if ( ! candidate . assigneeAgentId ) {
result . skipped += 1 ;
continue ;
}
if ( await isProductivityReviewDescendant ( candidate ) ) {
result . skipped += 1 ;
continue ;
}
if ( await findRecentResolvedProductivityReview ( candidate . companyId , candidate . id , thresholds , now ) ) {
result . snoozed += 1 ;
continue ;
}
const sourceAgent = await getAgent ( candidate . assigneeAgentId ) ;
if ( ! sourceAgent || sourceAgent . companyId !== candidate . companyId ) {
result . skipped += 1 ;
continue ;
}
const evidence = await collectEvidence ( candidate , sourceAgent , thresholds , now ) ;
if ( ! evidence ) {
result . skipped += 1 ;
continue ;
}
let prefix = prefixCache . get ( candidate . companyId ) ;
if ( ! prefix ) {
prefix = await getCompanyIssuePrefix ( candidate . companyId ) ;
prefixCache . set ( candidate . companyId , prefix ) ;
}
try {
[codex] Bound productivity review recovery loops (#4948)
## Thinking Path
> - Paperclip orchestrates AI agents for zero-human companies.
> - The heartbeat/productivity review subsystem detects when assigned
work is likely stuck or churning.
> - Productivity reviews are useful, but repeated reconciliation can
create noisy refresh comments or repeated review issues around the same
source issue.
> - That makes manager follow-up harder because the signal can get
buried under duplicate review activity.
> - This pull request bounds productivity review refreshes and creation
loops while preserving the existing escalation path.
> - The benefit is a quieter recovery loop that still surfaces stuck or
high-churn work for manager attention.
## What Changed
- Added refresh throttling for open productivity review issues,
including a one-hour default interval and a maximum of three refresh
comments per open review.
- Added a rolling 24-hour creation cap so completed/closed reviews
cannot immediately recreate review issues indefinitely for the same
source issue.
- Excluded cancelled productivity reviews from the creation cap so
manager cancellations do not silently suppress future legitimate
reviews.
- Preserved productivity review timestamps in deterministic test paths
and added targeted coverage for immediate refresh suppression, refresh
caps, creation caps, and cancelled-review exclusion.
## Verification
- `pnpm run preflight:workspace-links && pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- `pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- Greptile Review: 5/5 on commit
`bcf25832d0ffae25890b2ee7eed112d1c2d114fe` with review threads resolved.
- GitHub PR checks passed on the latest head: `policy`, `verify`, `e2e`,
`Greptile Review`, and `security/snyk (cryppadotta)`.
- Verified the branch is rebased onto `public-gh/master` with no
conflicts.
- Verified the diff does not include `pnpm-lock.yaml`, database schema
changes, or migrations.
## Risks
- Low-to-medium risk: this changes automation cadence for productivity
reviews. A truly stuck issue may receive fewer repeated refresh
comments, but the original review issue remains open and assigned for
manager action.
- No migration risk: this is server logic and tests only.
> Checked [`ROADMAP.md`](ROADMAP.md) for overlapping planned core work;
this is a targeted recovery-loop fix and does not add a new roadmap
feature.
## Model Used
- OpenAI Codex coding agent, GPT-5 model family, tool-using software
engineering mode. Exact context window is not exposed in this runtime.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots (not applicable; server-only change)
- [x] I have updated relevant documentation to reflect my changes (not
applicable; no user-facing docs or commands changed)
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-05-01 08:32:04 -05:00
const outcome = await createOrUpdateReview ( evidence , { prefix , thresholds } ) ;
[codex] Split backend control-plane QoL slice (#4700)
## Thinking Path
> - Paperclip is the control plane for autonomous AI companies, so
backend task ownership, recovery, review visibility, and company-scoped
limits need to stay enforceable without UI-only coupling.
> - Closed PR #4692 bundled those backend changes with UI workflow,
docs, skills, workflow, and lockfile churn.
> - PAP-2694 asks for a clean backend/control-plane slice from that
closed branch.
> - This branch starts from current `master` and mines only the `cli`,
`packages/db`, `packages/shared`, and `server` contracts/tests needed
for the backend behavior.
> - It explicitly excludes UI workflow/performance work,
`.github/workflows/pr.yml`, `pnpm-lock.yaml`, docs, skills,
package-script, adapter UI build-config, and perf fixture script
changes; the only UI files are fixture/test updates required by the
tightened shared `Company` contract.
> - The benefit is a smaller reviewable PR that preserves the
control-plane fixes while staying under Greptile s 100-file review
limit.
## What Changed
- Added company-scoped attachment-size limits through DB
schema/migrations, shared company portability contracts, CLI
import/export coverage, and server attachment upload enforcement.
- Added productivity review service/API behavior for no-comment streak,
long-active, and high-churn review issues, including request-depth
clamping and issue summary exposure.
- Hardened issue ownership and recovery/control-plane paths: peer-agent
mutation denial, issue tree pause/resume behavior, stranded recovery
origins, and related activity/test coverage.
- Preserved related backend contract updates for routine timestamp
variables and managed agent instruction bundles because they live in
shared/server contracts from the source branch.
- Addressed Greptile feedback by making `Company.attachmentMaxBytes`
non-optional, simplifying review request-depth clamping, fixing the
migration final newline, and enforcing the process-level attachment cap
as the final ceiling for uploads.
- Added minimal company fixtures needed for repo-wide typecheck/build
and kept the PR to 66 changed files with forbidden/non-slice paths
excluded.
## Verification
- `pnpm install --frozen-lockfile`
- `git diff --check origin/master..HEAD`
- `git diff --name-only origin/master..HEAD | wc -l` -> 66 files
- `git diff --name-only origin/master..HEAD -- .github/workflows/pr.yml
pnpm-lock.yaml package.json doc skills .agents scripts
packages/adapters` -> no output
- `pnpm exec vitest run --config vitest.config.ts
packages/shared/src/validators/issue.test.ts
packages/shared/src/routine-variables.test.ts
packages/shared/src/adapter-types.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
cli/src/__tests__/company.test.ts
server/src/__tests__/productivity-review-service.test.ts
server/src/__tests__/issue-tree-control-service.test.ts
server/src/__tests__/issue-tree-control-routes.test.ts
server/src/__tests__/issue-agent-mutation-ownership-routes.test.ts
server/src/__tests__/issue-attachment-routes.test.ts
server/src/__tests__/heartbeat-process-recovery.test.ts
server/src/__tests__/issues-service.test.ts` -> 12 files, 147 tests
passed
- `pnpm exec vitest run --config vitest.config.ts
cli/src/__tests__/company-delete.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
server/src/__tests__/productivity-review-service.test.ts` -> 3 files, 18
tests passed
- `pnpm exec vitest run --config vitest.config.ts
server/src/__tests__/issue-attachment-routes.test.ts` -> 1 file, 6 tests
passed
- `pnpm --filter @paperclipai/db typecheck && pnpm --filter
@paperclipai/shared typecheck && pnpm --filter @paperclipai/server
typecheck && pnpm --filter paperclipai typecheck`
- `pnpm --filter @paperclipai/server typecheck`
- `pnpm --filter @paperclipai/ui typecheck && pnpm --filter
@paperclipai/ui build`
## Risks
- Includes migrations `0073_shiny_salo.sql` and
`0074_striped_genesis.sql`; merge ordering matters if another PR adds
migrations first.
- This is intentionally backend-only apart from fixture/test updates
forced by shared type correctness; UI affordances from PR #4692 are not
present here and should land in separate UI slices.
- The worktree install emitted plugin SDK bin-link warnings for unbuilt
plugin packages, but the targeted tests and package typechecks completed
successfully.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected; check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex, GPT-5 coding agent, tool-enabled terminal/GitHub
workflow. Exact runtime context window was not exposed by the harness.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-04-28 16:46:45 -05:00
if ( outcome . kind === "created" ) result . created += 1 ;
else if ( outcome . kind === "updated" ) result . updated += 1 ;
[codex] Bound productivity review recovery loops (#4948)
## Thinking Path
> - Paperclip orchestrates AI agents for zero-human companies.
> - The heartbeat/productivity review subsystem detects when assigned
work is likely stuck or churning.
> - Productivity reviews are useful, but repeated reconciliation can
create noisy refresh comments or repeated review issues around the same
source issue.
> - That makes manager follow-up harder because the signal can get
buried under duplicate review activity.
> - This pull request bounds productivity review refreshes and creation
loops while preserving the existing escalation path.
> - The benefit is a quieter recovery loop that still surfaces stuck or
high-churn work for manager attention.
## What Changed
- Added refresh throttling for open productivity review issues,
including a one-hour default interval and a maximum of three refresh
comments per open review.
- Added a rolling 24-hour creation cap so completed/closed reviews
cannot immediately recreate review issues indefinitely for the same
source issue.
- Excluded cancelled productivity reviews from the creation cap so
manager cancellations do not silently suppress future legitimate
reviews.
- Preserved productivity review timestamps in deterministic test paths
and added targeted coverage for immediate refresh suppression, refresh
caps, creation caps, and cancelled-review exclusion.
## Verification
- `pnpm run preflight:workspace-links && pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- `pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- Greptile Review: 5/5 on commit
`bcf25832d0ffae25890b2ee7eed112d1c2d114fe` with review threads resolved.
- GitHub PR checks passed on the latest head: `policy`, `verify`, `e2e`,
`Greptile Review`, and `security/snyk (cryppadotta)`.
- Verified the branch is rebased onto `public-gh/master` with no
conflicts.
- Verified the diff does not include `pnpm-lock.yaml`, database schema
changes, or migrations.
## Risks
- Low-to-medium risk: this changes automation cadence for productivity
reviews. A truly stuck issue may receive fewer repeated refresh
comments, but the original review issue remains open and assigned for
manager action.
- No migration risk: this is server logic and tests only.
> Checked [`ROADMAP.md`](ROADMAP.md) for overlapping planned core work;
this is a targeted recovery-loop fix and does not add a new roadmap
feature.
## Model Used
- OpenAI Codex coding agent, GPT-5 model family, tool-using software
engineering mode. Exact context window is not exposed in this runtime.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots (not applicable; server-only change)
- [x] I have updated relevant documentation to reflect my changes (not
applicable; no user-facing docs or commands changed)
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-05-01 08:32:04 -05:00
else if ( outcome . kind === "creation_capped" ) result . creationCapped += 1 ;
[codex] Split backend control-plane QoL slice (#4700)
## Thinking Path
> - Paperclip is the control plane for autonomous AI companies, so
backend task ownership, recovery, review visibility, and company-scoped
limits need to stay enforceable without UI-only coupling.
> - Closed PR #4692 bundled those backend changes with UI workflow,
docs, skills, workflow, and lockfile churn.
> - PAP-2694 asks for a clean backend/control-plane slice from that
closed branch.
> - This branch starts from current `master` and mines only the `cli`,
`packages/db`, `packages/shared`, and `server` contracts/tests needed
for the backend behavior.
> - It explicitly excludes UI workflow/performance work,
`.github/workflows/pr.yml`, `pnpm-lock.yaml`, docs, skills,
package-script, adapter UI build-config, and perf fixture script
changes; the only UI files are fixture/test updates required by the
tightened shared `Company` contract.
> - The benefit is a smaller reviewable PR that preserves the
control-plane fixes while staying under Greptile s 100-file review
limit.
## What Changed
- Added company-scoped attachment-size limits through DB
schema/migrations, shared company portability contracts, CLI
import/export coverage, and server attachment upload enforcement.
- Added productivity review service/API behavior for no-comment streak,
long-active, and high-churn review issues, including request-depth
clamping and issue summary exposure.
- Hardened issue ownership and recovery/control-plane paths: peer-agent
mutation denial, issue tree pause/resume behavior, stranded recovery
origins, and related activity/test coverage.
- Preserved related backend contract updates for routine timestamp
variables and managed agent instruction bundles because they live in
shared/server contracts from the source branch.
- Addressed Greptile feedback by making `Company.attachmentMaxBytes`
non-optional, simplifying review request-depth clamping, fixing the
migration final newline, and enforcing the process-level attachment cap
as the final ceiling for uploads.
- Added minimal company fixtures needed for repo-wide typecheck/build
and kept the PR to 66 changed files with forbidden/non-slice paths
excluded.
## Verification
- `pnpm install --frozen-lockfile`
- `git diff --check origin/master..HEAD`
- `git diff --name-only origin/master..HEAD | wc -l` -> 66 files
- `git diff --name-only origin/master..HEAD -- .github/workflows/pr.yml
pnpm-lock.yaml package.json doc skills .agents scripts
packages/adapters` -> no output
- `pnpm exec vitest run --config vitest.config.ts
packages/shared/src/validators/issue.test.ts
packages/shared/src/routine-variables.test.ts
packages/shared/src/adapter-types.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
cli/src/__tests__/company.test.ts
server/src/__tests__/productivity-review-service.test.ts
server/src/__tests__/issue-tree-control-service.test.ts
server/src/__tests__/issue-tree-control-routes.test.ts
server/src/__tests__/issue-agent-mutation-ownership-routes.test.ts
server/src/__tests__/issue-attachment-routes.test.ts
server/src/__tests__/heartbeat-process-recovery.test.ts
server/src/__tests__/issues-service.test.ts` -> 12 files, 147 tests
passed
- `pnpm exec vitest run --config vitest.config.ts
cli/src/__tests__/company-delete.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
server/src/__tests__/productivity-review-service.test.ts` -> 3 files, 18
tests passed
- `pnpm exec vitest run --config vitest.config.ts
server/src/__tests__/issue-attachment-routes.test.ts` -> 1 file, 6 tests
passed
- `pnpm --filter @paperclipai/db typecheck && pnpm --filter
@paperclipai/shared typecheck && pnpm --filter @paperclipai/server
typecheck && pnpm --filter paperclipai typecheck`
- `pnpm --filter @paperclipai/server typecheck`
- `pnpm --filter @paperclipai/ui typecheck && pnpm --filter
@paperclipai/ui build`
## Risks
- Includes migrations `0073_shiny_salo.sql` and
`0074_striped_genesis.sql`; merge ordering matters if another PR adds
migrations first.
- This is intentionally backend-only apart from fixture/test updates
forced by shared type correctness; UI affordances from PR #4692 are not
present here and should land in separate UI slices.
- The worktree install emitted plugin SDK bin-link warnings for unbuilt
plugin packages, but the targeted tests and package typechecks completed
successfully.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected; check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex, GPT-5 coding agent, tool-enabled terminal/GitHub
workflow. Exact runtime context window was not exposed by the harness.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-04-28 16:46:45 -05:00
else result . existing += 1 ;
[codex] Bound productivity review recovery loops (#4948)
## Thinking Path
> - Paperclip orchestrates AI agents for zero-human companies.
> - The heartbeat/productivity review subsystem detects when assigned
work is likely stuck or churning.
> - Productivity reviews are useful, but repeated reconciliation can
create noisy refresh comments or repeated review issues around the same
source issue.
> - That makes manager follow-up harder because the signal can get
buried under duplicate review activity.
> - This pull request bounds productivity review refreshes and creation
loops while preserving the existing escalation path.
> - The benefit is a quieter recovery loop that still surfaces stuck or
high-churn work for manager attention.
## What Changed
- Added refresh throttling for open productivity review issues,
including a one-hour default interval and a maximum of three refresh
comments per open review.
- Added a rolling 24-hour creation cap so completed/closed reviews
cannot immediately recreate review issues indefinitely for the same
source issue.
- Excluded cancelled productivity reviews from the creation cap so
manager cancellations do not silently suppress future legitimate
reviews.
- Preserved productivity review timestamps in deterministic test paths
and added targeted coverage for immediate refresh suppression, refresh
caps, creation caps, and cancelled-review exclusion.
## Verification
- `pnpm run preflight:workspace-links && pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- `pnpm exec vitest run
server/src/__tests__/productivity-review-service.test.ts`
- Greptile Review: 5/5 on commit
`bcf25832d0ffae25890b2ee7eed112d1c2d114fe` with review threads resolved.
- GitHub PR checks passed on the latest head: `policy`, `verify`, `e2e`,
`Greptile Review`, and `security/snyk (cryppadotta)`.
- Verified the branch is rebased onto `public-gh/master` with no
conflicts.
- Verified the diff does not include `pnpm-lock.yaml`, database schema
changes, or migrations.
## Risks
- Low-to-medium risk: this changes automation cadence for productivity
reviews. A truly stuck issue may receive fewer repeated refresh
comments, but the original review issue remains open and assigned for
manager action.
- No migration risk: this is server logic and tests only.
> Checked [`ROADMAP.md`](ROADMAP.md) for overlapping planned core work;
this is a targeted recovery-loop fix and does not add a new roadmap
feature.
## Model Used
- OpenAI Codex coding agent, GPT-5 model family, tool-using software
engineering mode. Exact context window is not exposed in this runtime.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots (not applicable; server-only change)
- [x] I have updated relevant documentation to reflect my changes (not
applicable; no user-facing docs or commands changed)
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-05-01 08:32:04 -05:00
if ( outcome . reviewIssueId ) result . reviewIssueIds . push ( outcome . reviewIssueId ) ;
[codex] Split backend control-plane QoL slice (#4700)
## Thinking Path
> - Paperclip is the control plane for autonomous AI companies, so
backend task ownership, recovery, review visibility, and company-scoped
limits need to stay enforceable without UI-only coupling.
> - Closed PR #4692 bundled those backend changes with UI workflow,
docs, skills, workflow, and lockfile churn.
> - PAP-2694 asks for a clean backend/control-plane slice from that
closed branch.
> - This branch starts from current `master` and mines only the `cli`,
`packages/db`, `packages/shared`, and `server` contracts/tests needed
for the backend behavior.
> - It explicitly excludes UI workflow/performance work,
`.github/workflows/pr.yml`, `pnpm-lock.yaml`, docs, skills,
package-script, adapter UI build-config, and perf fixture script
changes; the only UI files are fixture/test updates required by the
tightened shared `Company` contract.
> - The benefit is a smaller reviewable PR that preserves the
control-plane fixes while staying under Greptile s 100-file review
limit.
## What Changed
- Added company-scoped attachment-size limits through DB
schema/migrations, shared company portability contracts, CLI
import/export coverage, and server attachment upload enforcement.
- Added productivity review service/API behavior for no-comment streak,
long-active, and high-churn review issues, including request-depth
clamping and issue summary exposure.
- Hardened issue ownership and recovery/control-plane paths: peer-agent
mutation denial, issue tree pause/resume behavior, stranded recovery
origins, and related activity/test coverage.
- Preserved related backend contract updates for routine timestamp
variables and managed agent instruction bundles because they live in
shared/server contracts from the source branch.
- Addressed Greptile feedback by making `Company.attachmentMaxBytes`
non-optional, simplifying review request-depth clamping, fixing the
migration final newline, and enforcing the process-level attachment cap
as the final ceiling for uploads.
- Added minimal company fixtures needed for repo-wide typecheck/build
and kept the PR to 66 changed files with forbidden/non-slice paths
excluded.
## Verification
- `pnpm install --frozen-lockfile`
- `git diff --check origin/master..HEAD`
- `git diff --name-only origin/master..HEAD | wc -l` -> 66 files
- `git diff --name-only origin/master..HEAD -- .github/workflows/pr.yml
pnpm-lock.yaml package.json doc skills .agents scripts
packages/adapters` -> no output
- `pnpm exec vitest run --config vitest.config.ts
packages/shared/src/validators/issue.test.ts
packages/shared/src/routine-variables.test.ts
packages/shared/src/adapter-types.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
cli/src/__tests__/company.test.ts
server/src/__tests__/productivity-review-service.test.ts
server/src/__tests__/issue-tree-control-service.test.ts
server/src/__tests__/issue-tree-control-routes.test.ts
server/src/__tests__/issue-agent-mutation-ownership-routes.test.ts
server/src/__tests__/issue-attachment-routes.test.ts
server/src/__tests__/heartbeat-process-recovery.test.ts
server/src/__tests__/issues-service.test.ts` -> 12 files, 147 tests
passed
- `pnpm exec vitest run --config vitest.config.ts
cli/src/__tests__/company-delete.test.ts
cli/src/__tests__/company-import-export-e2e.test.ts
server/src/__tests__/productivity-review-service.test.ts` -> 3 files, 18
tests passed
- `pnpm exec vitest run --config vitest.config.ts
server/src/__tests__/issue-attachment-routes.test.ts` -> 1 file, 6 tests
passed
- `pnpm --filter @paperclipai/db typecheck && pnpm --filter
@paperclipai/shared typecheck && pnpm --filter @paperclipai/server
typecheck && pnpm --filter paperclipai typecheck`
- `pnpm --filter @paperclipai/server typecheck`
- `pnpm --filter @paperclipai/ui typecheck && pnpm --filter
@paperclipai/ui build`
## Risks
- Includes migrations `0073_shiny_salo.sql` and
`0074_striped_genesis.sql`; merge ordering matters if another PR adds
migrations first.
- This is intentionally backend-only apart from fixture/test updates
forced by shared type correctness; UI affordances from PR #4692 are not
present here and should land in separate UI slices.
- The worktree install emitted plugin SDK bin-link warnings for unbuilt
plugin packages, but the targeted tests and package typechecks completed
successfully.
> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected; check the roadmap
first. See `CONTRIBUTING.md`.
## Model Used
- OpenAI Codex, GPT-5 coding agent, tool-enabled terminal/GitHub
workflow. Exact runtime context window was not exposed by the harness.
## Checklist
- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-04-28 16:46:45 -05:00
} catch ( err ) {
result . failed += 1 ;
result . failedIssueIds . push ( candidate . id ) ;
logger . warn (
{
err ,
companyId : candidate.companyId ,
issueId : candidate.id ,
requestDepth : candidate.requestDepth ,
} ,
"productivity review reconciliation skipped malformed candidate" ,
) ;
}
}
return result ;
}
async function isProductivityReviewContinuationHoldActive ( input : {
companyId : string ;
issueId : string ;
agentId : string ;
now? : Date ;
thresholds? : Partial < ProductivityReviewThresholds > ;
} ) {
const now = input . now ? ? new Date ( ) ;
const thresholds = buildThresholds ( input . thresholds ) ;
const [ sourceIssue , sourceAgent , openReview ] = await Promise . all ( [
db
. select ( )
. from ( issues )
. where ( and ( eq ( issues . companyId , input . companyId ) , eq ( issues . id , input . issueId ) ) )
. then ( ( rows ) = > rows [ 0 ] ? ? null ) ,
getAgent ( input . agentId ) ,
findOpenProductivityReview ( input . companyId , input . issueId ) ,
] ) ;
if ( ! sourceIssue || ! sourceAgent || ! openReview ) return { held : false as const } ;
if ( sourceAgent . companyId !== input . companyId ) return { held : false as const } ;
const evidence = await collectEvidence ( sourceIssue , sourceAgent , thresholds , now ) ;
if ( ! evidence || ! isSoftStopTrigger ( evidence . trigger ) ) return { held : false as const } ;
return {
held : true as const ,
reviewIssueId : openReview.id ,
reviewIdentifier : openReview.identifier ,
trigger : evidence.trigger ,
reason : evidence.triggerReasons.join ( "; " ) ,
} ;
}
async function recordContinuationHold ( input : {
companyId : string ;
issueId : string ;
runId : string ;
agentId : string ;
reviewIssueId : string ;
trigger : ProductivityReviewTrigger ;
reason : string ;
} ) {
await logActivity ( db , {
companyId : input.companyId ,
actorType : "system" ,
actorId : "system" ,
agentId : input.agentId ,
runId : input.runId ,
action : "issue.productivity_review_continuation_held" ,
entityType : "issue" ,
entityId : input.issueId ,
details : {
source : "productivity_review.continuation_hold" ,
reviewIssueId : input.reviewIssueId ,
trigger : input.trigger ,
reason : input.reason ,
} ,
} ) ;
}
return {
reconcileProductivityReviews ,
isProductivityReviewContinuationHoldActive ,
recordContinuationHold ,
} ;
}