[codex] Stabilize tests and local maintenance assets (#4423)

## Thinking Path

> - Paperclip orchestrates AI agents for zero-human companies
> - A fast-moving control plane needs stable local tests and repeatable
local maintenance tools so contributors can safely split and review work
> - Several route suites needed stronger isolation, Codex manual model
selection needed a faster-mode option, and local browser cleanup missed
Playwright's headless shell binary
> - Storybook static output also needed to be preserved as a generated
review artifact from the working branch
> - This pull request groups the test/local-dev maintenance pieces so
they can be reviewed separately from product runtime changes
> - The benefit is more predictable contributor verification and cleaner
local maintenance without mixing these changes into feature PRs

## What Changed

- Added stable Vitest runner support and serialized route/authz test
isolation.
- Fixed workspace runtime authz route mocks and stabilized
Claude/company-import related assertions.
- Allowed Codex fast mode for manually selected models.
- Broadened the agent browser cleanup script to detect
`chrome-headless-shell` as well as Chrome for Testing.
- Preserved generated Storybook static output from the source branch.

## Verification

- `pnpm exec vitest run
src/__tests__/workspace-runtime-routes-authz.test.ts
src/__tests__/claude-local-execute.test.ts --config vitest.config.ts`
from `server/` passed: 2 files, 19 tests.
- `pnpm exec vitest run src/server/codex-args.test.ts --config
vitest.config.ts` from `packages/adapters/codex-local/` passed: 1 file,
3 tests.
- `bash -n scripts/kill-agent-browsers.sh &&
scripts/kill-agent-browsers.sh --dry` passed; dry-run detected
`chrome-headless-shell` processes without killing them.
- `test -f ui/storybook-static/index.html && test -f
ui/storybook-static/assets/forms-editors.stories-Dry7qwx2.js` passed.
- `git diff --check public-gh/master..pap-2228-test-local-maintenance --
. ':(exclude)ui/storybook-static'` passed.
- `pnpm exec vitest run
cli/src/__tests__/company-import-export-e2e.test.ts --config
cli/vitest.config.ts` did not complete in the isolated split worktree
because `paperclipai run` exited during build prep with `TS2688: Cannot
find type definition file for 'react'`; this appears to be caused by the
worktree dependency symlink setup, not the code under test.
- Confirmed this PR does not include `pnpm-lock.yaml`.

## Risks

- Medium risk: the stable Vitest runner changes how route/authz tests
are scheduled.
- Generated `ui/storybook-static` files are large and contain minified
third-party output; `git diff --check` reports whitespace inside those
generated assets, so reviewers may choose to drop or regenerate that
artifact before merge.
- No database migrations.

> 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 based on GPT-5, with shell, git, Paperclip
API, and GitHub CLI tool use in the local Paperclip workspace.

## 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

Note: screenshot checklist item is not applicable to source UI behavior;
the included Storybook static output is generated artifact preservation
from the source branch.

---------

Co-authored-by: Paperclip <noreply@paperclip.ing>
This commit is contained in:
Dotta 2026-04-24 15:11:42 -05:00 committed by GitHub
parent 70679a3321
commit 9a8d219949
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
56 changed files with 1250 additions and 763 deletions

View file

@ -4,7 +4,23 @@ export const DEFAULT_CODEX_LOCAL_MODEL = "gpt-5.3-codex";
export const DEFAULT_CODEX_LOCAL_BYPASS_APPROVALS_AND_SANDBOX = true;
export const CODEX_LOCAL_FAST_MODE_SUPPORTED_MODELS = ["gpt-5.4"] as const;
function normalizeModelId(model: string | null | undefined): string {
return typeof model === "string" ? model.trim() : "";
}
export function isCodexLocalKnownModel(model: string | null | undefined): boolean {
const normalizedModel = normalizeModelId(model);
if (!normalizedModel) return false;
return models.some((entry) => entry.id === normalizedModel);
}
export function isCodexLocalManualModel(model: string | null | undefined): boolean {
const normalizedModel = normalizeModelId(model);
return Boolean(normalizedModel) && !isCodexLocalKnownModel(normalizedModel);
}
export function isCodexLocalFastModeSupported(model: string | null | undefined): boolean {
if (isCodexLocalManualModel(model)) return true;
const normalizedModel = typeof model === "string" ? model.trim() : "";
return CODEX_LOCAL_FAST_MODE_SUPPORTED_MODELS.includes(
normalizedModel as (typeof CODEX_LOCAL_FAST_MODE_SUPPORTED_MODELS)[number],
@ -35,7 +51,7 @@ Core fields:
- modelReasoningEffort (string, optional): reasoning effort override (minimal|low|medium|high|xhigh) passed via -c model_reasoning_effort=...
- promptTemplate (string, optional): run prompt template
- search (boolean, optional): run codex with --search
- fastMode (boolean, optional): enable Codex Fast mode; currently supported on GPT-5.4 only and consumes credits faster
- fastMode (boolean, optional): enable Codex Fast mode; supported on GPT-5.4 and passed through for manual model IDs
- dangerouslyBypassApprovalsAndSandbox (boolean, optional): run with bypass flag
- command (string, optional): defaults to "codex"
- extraArgs (string[], optional): additional CLI args
@ -54,6 +70,6 @@ Notes:
- Paperclip injects desired local skills into the effective CODEX_HOME/skills/ directory at execution time so Codex can discover "$paperclip" and related skills without polluting the project working directory. In managed-home mode (the default) this is ~/.paperclip/instances/<id>/companies/<companyId>/codex-home/skills/; when CODEX_HOME is explicitly overridden in adapter config, that override is used instead.
- Unless explicitly overridden in adapter config, Paperclip runs Codex with a per-company managed CODEX_HOME under the active Paperclip instance and seeds auth/config from the shared Codex home (the CODEX_HOME env var, when set, or ~/.codex).
- Some model/tool combinations reject certain effort levels (for example minimal with web search enabled).
- Fast mode is currently supported on GPT-5.4 only. When enabled, Paperclip applies \`service_tier="fast"\` and \`features.fast_mode=true\`.
- Fast mode is supported on GPT-5.4 and manual model IDs. When enabled for those models, Paperclip applies \`service_tier="fast"\` and \`features.fast_mode=true\`.
- When Paperclip realizes a workspace/runtime for a run, it injects PAPERCLIP_WORKSPACE_* and PAPERCLIP_RUNTIME_* env vars for agent-side tooling.
`;

View file

@ -26,6 +26,28 @@ describe("buildCodexExecArgs", () => {
]);
});
it("enables Codex fast mode overrides for manual models", () => {
const result = buildCodexExecArgs({
model: "gpt-5.5",
fastMode: true,
});
expect(result.fastModeRequested).toBe(true);
expect(result.fastModeApplied).toBe(true);
expect(result.fastModeIgnoredReason).toBeNull();
expect(result.args).toEqual([
"exec",
"--json",
"--model",
"gpt-5.5",
"-c",
'service_tier="fast"',
"-c",
"features.fast_mode=true",
"-",
]);
});
it("ignores fast mode for unsupported models", () => {
const result = buildCodexExecArgs({
model: "gpt-5.3-codex",
@ -34,7 +56,9 @@ describe("buildCodexExecArgs", () => {
expect(result.fastModeRequested).toBe(true);
expect(result.fastModeApplied).toBe(false);
expect(result.fastModeIgnoredReason).toContain("currently only supported on gpt-5.4");
expect(result.fastModeIgnoredReason).toContain(
"currently only supported on gpt-5.4 or manually configured model IDs",
);
expect(result.args).toEqual([
"exec",
"--json",

View file

@ -25,7 +25,7 @@ function asRecord(value: unknown): Record<string, unknown> {
}
function formatFastModeSupportedModels(): string {
return CODEX_LOCAL_FAST_MODE_SUPPORTED_MODELS.join(", ");
return `${CODEX_LOCAL_FAST_MODE_SUPPORTED_MODELS.join(", ")} or manually configured model IDs`;
}
export function buildCodexExecArgs(

View file

@ -146,7 +146,7 @@ export async function testEnvironment(
code: "codex_fast_mode_unsupported_model",
level: "warn",
message: execArgs.fastModeIgnoredReason,
hint: "Switch the agent model to GPT-5.4 to enable Codex Fast mode.",
hint: "Switch the agent model to GPT-5.4 or enter a manual model ID to enable Codex Fast mode.",
});
}