> _Stacked on top of #5685 (Harden remote sandbox runtime). Diff against
master includes commits from earlier PRs in the stack — review focuses
on the new commit only._
## Thinking Path
> - Paperclip orchestrates AI agents for zero-human companies
> - The cursor-local adapter wraps the Cursor Agent CLI so a Paperclip
workflow can drive it inside a sandbox
> - When the adapter runs in a remote sandbox, the Cursor Agent CLI
installs under `$HOME/.local/bin/cursor-agent` (or wherever
`$XDG_BIN_HOME` points), not on the global PATH
> - The existing post-install resolution assumed `cursor-agent` would
resolve via the sandbox's login shell PATH after `npm install -g`, which
fails on sandboxes where the install lands in a user-prefixed directory
that isn't on PATH at probe time
> - This pull request resolves the agent CLI from the cursor binary's
own directory (`dirname "$(command -v cursor)"`) so the install probe
and execute path agree on a real binary location
> - The benefit is that cursor-local works correctly on any sandbox
provider where `npm install` lands in a user-prefixed directory
## What Changed
- `packages/adapters/cursor-local/src/server/remote-command.ts`: resolve
the cursor-agent binary from the cursor bin directory after install,
instead of relying on PATH.
- `packages/adapters/cursor-local/src/server/test.ts`: corresponding
probe tweak.
- `packages/adapters/cursor-local/src/server/test.test.ts` (new) +
`remote-command.test.ts`: focused coverage that exercises the install +
resolve path against a sandbox runner that places the binary in a
user-prefixed directory.
## Verification
- `pnpm exec vitest run --no-coverage
packages/adapters/cursor-local/src/server/test.test.ts
packages/adapters/cursor-local/src/server/remote-command.test.ts
packages/adapters/cursor-local/src/server/execute.test.ts`
All passing locally.
## Risks
- Local cursor-local runs are unaffected — the resolution change only
kicks in for the sandbox install path.
- Low risk; isolated to one adapter.
## Model Used
- Provider: Anthropic
- Model: Claude Opus 4.7 (1M context)
- Capabilities used: tool use (Read/Edit/Bash), no code execution beyond
local repo commands
## 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
- [ ] If this change affects the UI, I have included before/after
screenshots — N/A, no UI change
- [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>
## Thinking Path
> - Paperclip orchestrates AI agents for zero-human companies
> - The Cursor adapter spawns the Cursor CLI against local, SSH, and
sandbox execution targets; on a fresh sandbox lease, it has to resolve
where Cursor was installed
> - The previous resolver only looked for `~/.local/bin/cursor-agent`
even though the official installer (and the adapter's own
`SANDBOX_INSTALL_COMMAND`) sometimes lays the binary down as
`~/.local/bin/agent`, so a sandbox where the install ran successfully
would still fail to find the CLI
> - This pull request lets the resolver accept either basename and lets
the caller pass an optional `remoteSystemHomeDirHint` so a probe doesn't
pay the cost of a remote `printf $HOME` round-trip when the home
directory is already known
> - The benefit is sandboxed Cursor runs find the binary that the
install actually produced, and runtime probes are cheaper when the home
dir is already resolved
## What Changed
- `packages/adapters/cursor-local/src/server/remote-command.ts`: accept
either `agent` or `cursor-agent` as the preferred basename; new optional
`remoteSystemHomeDirHint` short-circuits the home-dir probe
- `packages/adapters/cursor-local/src/server/execute.ts`: thread the
home-dir hint through, prefer the resolved binary path, and shift the
effective execution cwd to the per-run managed subdirectory once the
runtime is prepared
- New `remote-command.test.ts` and `execute.test.ts` cover both
basenames, the hint short-circuit, and the cwd shift
- `packages/adapters/cursor-local/src/index.ts`: update doc string to
reflect the broader resolution
- `execute.remote.test.ts` updated to expect the managed-subdirectory
cwd shape introduced by the cwd shift
## Verification
- `pnpm vitest run --no-coverage --project
@paperclipai/adapter-cursor-local` — 6/6 passing
- `pnpm typecheck` clean
- Manual: a fresh sandbox lease with `npm install -g …`-installed Cursor
(binary lands as `~/.local/bin/agent`) now runs cleanly through the
adapter
## Risks
Low. Resolver is strictly broader (matches a superset of paths);
existing setups with `~/.local/bin/cursor-agent` continue to work. The
home-dir hint is opt-in; callers that don't pass it get the existing
probe behavior. Cursor's effective execution cwd now matches the rest of
the adapters (per-run managed subdirectory) — sessions previously rooted
at the workspace root will land in the new subdirectory.
## Model Used
Claude Opus 4.7 (1M context)
## 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 — new tests cover
both basenames + hint short-circuit + cwd shift
- [x] If this change affects the UI, I have included before/after
screenshots — N/A (no UI)
- [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
---
> **Stacked PR.** Sits on top of #5445 (which sits on #5444). Cumulative
diff against `master` includes both of those PRs' content; the files
touched by *this* PR's commit are listed under "What Changed" above.
Will rebase onto `master` and force-push once the prerequisite PRs
merge.
> **Stacked PR.** Sits on top of the e2b sandbox chain — #5278 (stdin
staging) and #5279 (honest-resolvability + login-profiles). The
cumulative diff against `master` includes both of those PRs' content;
the files touched by *this* PR's commit are the new
`maybeRunSandboxInstallCommand` helper in
`packages/adapter-utils/src/execution-target.ts` and the per-adapter
`index.ts`/`server/test.ts`/`server/execute.ts` wiring under
`packages/adapters/{claude,codex,cursor,gemini,opencode,pi}-local/`. The
honest resolvability check from #5279 is what gives this PR's install
command a meaningful "did it actually land on PATH" follow-up.
## Thinking Path
> - Paperclip orchestrates AI agents for zero-human companies
> - Sandbox execution targets are ephemeral — each fresh lease starts
from a template image that may or may not have the agent CLIs
preinstalled
> - When a CLI isn't preinstalled, the resolvability probe fails at
`command -v` and the hello probe never runs
> - There's no shared mechanism for "before you probe or provision,
install the CLI on this sandbox"
> - This pull request adds a `SANDBOX_INSTALL_COMMAND` constant per
adapter and a `maybeRunSandboxInstallCommand` helper that runs it via
the existing sandbox login shell, captures structured output, and never
throws (so the resolvability + hello probe still run after); each
adapter's `test()` and `execute()` share the constant so the two
callsites can't drift
> - The benefit is a fresh sandbox lease without a preinstalled CLI now
installs it once via `sh -lc` before the resolvability probe and before
managed-runtime provisioning, with a uniform
`<adapter>_install_command_run` check on the test report
## What Changed
- `packages/adapter-utils/src/execution-target.ts`: add
`AdapterSandboxInstallCommandCheck` and `maybeRunSandboxInstallCommand`
(runs the install via existing sandbox shell, captures
exit/stdout/stderr, returns a structured info/warn check, never throws)
- Add `SANDBOX_INSTALL_COMMAND` to each adapter's `index.ts` so `test()`
and `execute()` share a single source of truth
- Wire each of the 6 affected adapter `testEnvironment()`s to call
`maybeRunSandboxInstallCommand` before
`ensureAdapterExecutionTargetCommandResolvable`
- Pass `installCommand: SANDBOX_INSTALL_COMMAND` through
`prepareAdapterExecutionTargetRuntime` in each adapter's `execute()`
- Per-adapter install commands use npm globals where possible so
binaries land on a PATH segment the template already exports:
- claude → `npm install -g @anthropic-ai/claude-code`
- codex → `npm install -g @openai/codex`
- cursor → `curl https://cursor.com/install -fsS | bash`
- gemini → `npm install -g @google/gemini-cli`
- opencode → `npm install -g opencode-ai`
- pi → `npm install -g @mariozechner/pi-coding-agent`
SSH and local targets ignore `installCommand` (SSH runtime takes no such
param; local short-circuits before runtime prep), so this is a no-op for
non-sandbox environments.
## Verification
- `pnpm typecheck` clean
- `pnpm vitest run --no-coverage --project @paperclipai/adapter-utils`
and per-adapter projects pass
- Manual sandbox matrix (claude, codex, cursor, gemini, opencode, pi) —
each goes `install_command_run → resolvable → hello_probe_passed` (Codex
and Pi land on `hello_probe_auth_required`, which is the
configured-credentials problem, not an install issue)
- SSH no-regression: SSH Claude still passes; the helper short-circuits
on non-sandbox targets
## Risks
Medium — adds a network/CPU cost (npm install / curl) on every fresh
sandbox lease. Cost is bounded (one-time per lease, typically tens of
seconds for npm globals), and the helper never throws so a failing
install still lets the report run resolvability and hello probes. If a
sandbox image already has the CLI, the install is an idempotent
reinstall.
## Model Used
Claude Opus 4.7 (1M context)
## 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 UI)
- [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
## Thinking Path
> - Paperclip is a control plane for autonomous AI companies, where
adapters are the boundary between the board, agents, and execution
runtimes.
> - Local adapters currently expose a primary runtime configuration, but
operators often need a cheaper model lane for routine or low-risk work.
> - That cheap lane has to stay adapter-owned: runtime profile settings
should not mutate the primary adapter config or bypass existing
auth/secret mediation.
> - Issue creation also needs an ergonomic way to request primary,
cheap, or custom model behavior for a selected assignee.
> - This pull request adds a first-class `cheap` model profile contract
across adapter capabilities, heartbeat config resolution, agent
configuration, and issue creation.
> - The benefit is cheaper task execution can be configured and
requested explicitly while preserving adapter boundaries, secret
handling, and audit visibility.
## What Changed
- Added adapter model-profile capability metadata and a `cheap` profile
contract for supported local adapters.
- Applied `runtimeConfig.modelProfiles.cheap.adapterConfig` during
heartbeat config resolution, including requested/applied/fallback run
metadata.
- Added agent configuration UI for cheap model profile settings without
writing those settings into primary `adapterConfig`.
- Added New Issue assignee model lane controls for Primary / Cheap /
Custom and request payload handling.
- Added run ledger profile badges and Storybook stories for the new
cheap-lane UI states.
- Added tests for validators, heartbeat model profile application,
permission/secret mediation, UI payload helpers, and run ledger
rendering.
- Added committed UI verification screenshots under
`docs/pr-screenshots/pap-2837/`.
- Addressed Greptile review feedback around cheap-profile defaults,
shared profile types, and fallback test data.
## Verification
Local:
- `pnpm exec vitest run packages/shared/src/validators/issue.test.ts
server/src/__tests__/adapter-registry.test.ts
server/src/__tests__/agent-permissions-routes.test.ts
server/src/__tests__/heartbeat-model-profile.test.ts
ui/src/components/IssueRunLedger.test.tsx
ui/src/lib/agent-config-patch.test.ts
ui/src/lib/issue-assignee-overrides.test.ts
ui/src/lib/new-agent-runtime-config.test.ts` — passed, 8 files / 103
tests.
- `pnpm exec vitest run ui/src/lib/new-agent-runtime-config.test.ts
ui/src/components/IssueRunLedger.test.tsx` — passed after
Greptile/rebase follow-up, 2 files / 17 tests.
- `pnpm --filter @paperclipai/ui typecheck` — passed after
Greptile/rebase follow-up.
- `pnpm -r typecheck` — passed.
- `pnpm build` — passed.
- `pnpm test:run` — did not complete successfully in this local
worktree: it stopped in pre-existing `@paperclipai/adapter-utils`
sandbox/SSH fixture suites outside this PR diff. Failures were 5s local
timeouts plus `git init -b` unsupported by this machine's Git 2.21.0.
The branch-specific targeted suites above passed.
- Branch was fetched/rebased onto `public-gh/master`; `git rev-list
--left-right --count public-gh/master...HEAD` reports `0 9`.
Remote PR checks on latest head
`e30bf399146451c86cee98ed528d51d33fa5af5a`:
- `policy` — passed.
- `verify` — passed.
- `e2e` — passed.
- `Greptile Review` — passed, confidence score 5/5; Greptile review
threads resolved.
- `security/snyk (cryppadotta)` — passed.
Screenshots:
- [New issue cheap lane
desktop](https://github.com/paperclipai/paperclip/blob/PAP-2837-plan-cheap-model-for-adapters-that-can-support-it/docs/pr-screenshots/pap-2837/newissue-cheap-desktop.png)
- [New issue custom lane
desktop](https://github.com/paperclipai/paperclip/blob/PAP-2837-plan-cheap-model-for-adapters-that-can-support-it/docs/pr-screenshots/pap-2837/newissue-custom-desktop.png)
- [New issue unsupported adapter
desktop](https://github.com/paperclipai/paperclip/blob/PAP-2837-plan-cheap-model-for-adapters-that-can-support-it/docs/pr-screenshots/pap-2837/newissue-unsupported-desktop.png)
- [Run ledger model profile badges
desktop](https://github.com/paperclipai/paperclip/blob/PAP-2837-plan-cheap-model-for-adapters-that-can-support-it/docs/pr-screenshots/pap-2837/runledger-profile-badges-desktop.png)
- Mobile variants are also in `docs/pr-screenshots/pap-2837/`.
## Risks
- Medium: heartbeat config mediation now merges runtime model profiles
into adapter configs, so adapter secret normalization and host-command
restrictions must keep covering nested config paths.
- Medium: the UI adds another issue creation choice; unsupported
adapters must keep hiding the cheap lane and preserve primary behavior.
- Low migration risk: no database migration is included.
> 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 using GPT-5-class reasoning with repo tool use
and command execution. Exact served model/context window was not exposed
by the 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
- [ ] 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>
Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com>
## Thinking Path
> - Paperclip orchestrates AI agents for zero-human companies
> - Agents can run inside sandboxed environments like E2B, or on remote
hosts via SSH
> - The cursor adapter needs to resolve `cursor-agent` inside sandbox
environments where it's installed in `~/.local/bin`
> - But when using the default `agent` command on a sandbox target, the
adapter didn't know to look in `~/.local/bin/cursor-agent`, causing
"command not found" failures
> - Additionally, repeated SSH runs failed because `git checkout` during
workspace sync conflicted with leftover `.paperclip-runtime` files from
previous runs
> - This PR adds sandbox-aware command resolution for cursor and fixes
the SSH workspace sync conflict
> - The benefit is cursor works in E2B sandboxes out of the box, and
repeated SSH runs don't fail on workspace sync
## What Changed
- `cursor-local`: Added `prepareCursorSandboxCommand` — on sandbox
targets, reads the remote `$HOME`, prepends `~/.local/bin` to PATH, and
prefers `~/.local/bin/cursor-agent` when the default command is
requested; tightened the sandbox command probe to validate the binary
exists before launching; preserves explicit custom command overrides
- `adapter-utils/ssh.ts`: Added `--force` to git checkout in SSH
workspace sync to handle `.paperclip-runtime` untracked file conflicts
from previous runs
## Verification
- `pnpm test` — all existing and new tests pass, including cursor
sandbox probe, sandbox execution, and custom command override tests
- `pnpm typecheck` — clean
- Manual: configure an E2B environment, run a cursor-local task, verify
it resolves cursor-agent from the sandbox install path
## Risks
- Low-medium. The `--force` flag on git checkout could discard
uncommitted changes in the remote workspace, but the workspace is
managed by Paperclip and should not contain user edits.
## Model Used
Codex GPT 5.4 high via Paperclip.
## 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
- [ ] 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
The --yolo flag bypasses interactive prompts more broadly than --trust.
Updated execute, test probe, docs, and test expectations.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Export ensureCursorSkillsInjected from the server entrypoint and add
a test for skill directory injection. Document the auto-inject and
auto-trust behaviors in the adapter notes.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
