fix: treat empty baseUrl as non-direct for OpenAI Responses by weiguangli-io · Pull Request #16930 · openclaw/openclaw

weiguangli-io · 2026-02-15T07:55:24Z

Summary

isDirectOpenAIBaseUrl() currently treats empty/falsy baseUrl as direct OpenAI, which makes the Responses wrapper force payload.store = true.

For models routed through non-direct providers with empty baseUrl metadata (including the reported openai-codex/gpt-5.3-codex path), this is incorrect and triggers upstream errors.

This PR changes empty baseUrl handling to return false (not direct OpenAI).

Changes

src/agents/pi-embedded-runner/extra-params.ts
- isDirectOpenAIBaseUrl("" | missing) now returns false.
src/agents/pi-embedded-runner-extraparams.e2e.test.ts
- added regression test that verifies empty baseUrl does not force store=true.

Testing

pnpm exec vitest run --config vitest.e2e.config.ts src/agents/pi-embedded-runner-extraparams.e2e.test.ts

AI-assisted

AI-assisted: yes. I manually reviewed and validated the final patch and tests.

Greptile Summary

Fixes a bug where isDirectOpenAIBaseUrl() treated empty/falsy baseUrl as direct OpenAI, causing the Responses API wrapper to force payload.store = true for models routed through non-direct providers (e.g., openai-codex/gpt-5.3-codex paths with empty baseUrl metadata). The fix changes the empty baseUrl case to return false, so only explicitly recognized OpenAI URLs (api.openai.com, chatgpt.com) trigger the store=true override.

src/agents/pi-embedded-runner/extra-params.ts: One-line change from return true to return false in the early-return guard of isDirectOpenAIBaseUrl. Models with explicit direct OpenAI baseUrl values are unaffected.
src/agents/pi-embedded-runner-extraparams.e2e.test.ts: Adds a regression test verifying that empty baseUrl does not force store=true.
.github/workflows/formal-conformance.yml: Adds continue-on-error: true to the PR comment step, preventing CI failure when the workflow lacks permissions to post comments (e.g., fork PRs).

Confidence Score: 5/5

This PR is safe to merge — it's a minimal, well-scoped bug fix with proper test coverage.
The change is a single-line fix (return true → return false) that corrects incorrect behavior for empty baseUrl values. The function is only called from one place (shouldForceResponsesStore), the change is covered by a new regression test, and the existing test suite confirms that direct OpenAI URLs still correctly force store=true. The CI workflow change is similarly low-risk — continue-on-error: true only affects an informational comment step.
No files require special attention.

_{Last reviewed commit: 04d8569}

_{(2/5) Greptile learns from your feedback when you react with thumbs up/down!}

weiguangli-io · 2026-02-15T07:59:30Z

Quick note on the current red check: formal_conformance failed because the workflow could not post its informational comment on a fork PR (Resource not accessible by integration, 403).

The code/test checks are green on this branch, and this failure appears to be workflow permission-related rather than a functional regression.

weiguangli-io · 2026-02-15T08:05:04Z

Pushed a small CI-only follow-up to address the red formal_conformance check.

Root cause was not functional code regression: the workflow’s informational PR comment step fails with 403 on fork PRs (Resource not accessible by integration), which marked the job red.

Change made:

.github/workflows/formal-conformance.yml
added continue-on-error: true on the informational comment step so permission-denied comment attempts do not fail the job.

This keeps drift detection/reporting behavior while preventing false-red CI on forks.

weiguangli-io · 2026-02-17T05:00:37Z

Quick check update on the current red CI:

checks (node, test, pnpm canvas:a2ui:bundle && pnpm test) failed on extensions/mattermost/src/mattermost/monitor-websocket.test.ts with a timeout in dispatches reaction events to the reaction handler.
The same run also showed Node OOM traces during the large test matrix (heap out of memory), which points to an environment/runtime instability rather than the PR diff.
checks-windows logs show a full test pass summary (335 passed, 1 skipped), but the overall job still exited non-zero from the aggregated run.

This PR only changes:

src/agents/pi-embedded-runner/extra-params.ts
src/agents/pi-embedded-runner-extraparams.e2e.test.ts

Given the failure signatures and touched paths, this looks unrelated to the PR changes. Could a maintainer rerun the failed checks once?

weiguangli-io · 2026-02-17T05:52:08Z

Quick follow-up data point:

I ran the previously failing Mattermost test locally on this branch:

pnpm vitest run extensions/mattermost/src/mattermost/monitor-websocket.test.ts

Result: pass (2 passed).

So the current CI red status still looks consistent with a runner/environment instability in that large matrix run rather than a deterministic regression in this PR.

, #22441, #25088) Landed from contributor PRs #16930, #22441, and #25088. Co-authored-by: liweiguang <[email protected]> Co-authored-by: EdwardWu7 <[email protected]> Co-authored-by: MoerAI <[email protected]>

steipete · 2026-02-27T16:16:14Z

Landed on main as 8da3a9a92d9c2c6db6fa6caa881e9310d1792e93.

Rewrite summary:

Kept fix: treat empty baseUrl as non-direct for OpenAI Responses #16930 behavior: empty baseUrl is treated as non-direct OpenAI Responses.
Unified with fix: force store for openai-responses providers #22441 + fix(agents): skip forced store=true when model declares supportsStore=false #25088 into one OpenAI Responses context wrapper.
Added automatic OpenAI server-side context_management compaction injection for compatible direct OpenAI models.
Added tests, docs, and one unified changelog entry.

SHA notes:

Original PR commit: f947032b8dc7d5e69555e2c09211572999f44b42
Landed commit: 8da3a9a92d9c2c6db6fa6caa881e9310d1792e93

Thanks @OiPunk.

…nclaw#16930, openclaw#22441, openclaw#25088) Landed from contributor PRs openclaw#16930, openclaw#22441, and openclaw#25088. Co-authored-by: liweiguang <[email protected]> Co-authored-by: EdwardWu7 <[email protected]> Co-authored-by: MoerAI <[email protected]>

…nclaw#16930, openclaw#22441, openclaw#25088) Landed from contributor PRs openclaw#16930, openclaw#22441, and openclaw#25088. Co-authored-by: liweiguang <[email protected]> Co-authored-by: EdwardWu7 <[email protected]> Co-authored-by: MoerAI <[email protected]> (cherry picked from commit c38be8f)

…nclaw#16930, openclaw#22441, openclaw#25088) Landed from contributor PRs openclaw#16930, openclaw#22441, and openclaw#25088. Co-authored-by: liweiguang <[email protected]> Co-authored-by: EdwardWu7 <[email protected]> Co-authored-by: MoerAI <[email protected]>

openclaw-barnacle bot added agents Agent runtime and tooling size: XS labels Feb 15, 2026

thewilloftheshadow force-pushed the main branch from bfc1ccb to f92900f Compare February 15, 2026 18:46

steipete closed this Feb 16, 2026

steipete reopened this Feb 17, 2026

openclaw-barnacle bot added size: S and removed size: XS labels Feb 17, 2026

fix: avoid forcing store for empty OpenAI baseUrl

f947032

weiguangli-io force-pushed the codex/openclaw-fix-empty-baseurl-store-gate branch from 7dfbfd2 to f947032 Compare February 17, 2026 09:00

openclaw-barnacle bot added size: XS and removed size: S labels Feb 17, 2026

This was referenced Feb 27, 2026

fix: force store for openai-responses providers #22441

Closed

fix(agents): skip forced store=true when model declares supportsStore=false #25088

Closed

steipete closed this Feb 27, 2026

gemini-code-assist bot mentioned this pull request Mar 1, 2026

chore(sync): replay rollup fixes on upstream/main MillionthOdin16/openclaw#111

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

fix: treat empty baseUrl as non-direct for OpenAI Responses#16930

fix: treat empty baseUrl as non-direct for OpenAI Responses#16930
weiguangli-io wants to merge 1 commit intoopenclaw:mainfrom
weiguangli-io:codex/openclaw-fix-empty-baseurl-store-gate

weiguangli-io commented Feb 15, 2026 •

edited by greptile-apps bot

Loading

Uh oh!

weiguangli-io commented Feb 15, 2026

Uh oh!

weiguangli-io commented Feb 15, 2026

Uh oh!

weiguangli-io commented Feb 17, 2026

Uh oh!

weiguangli-io commented Feb 17, 2026

Uh oh!

steipete commented Feb 27, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Conversation

weiguangli-io commented Feb 15, 2026 • edited by greptile-apps bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Changes

Testing

AI-assisted

Greptile Summary

Confidence Score: 5/5

Uh oh!

weiguangli-io commented Feb 15, 2026

Uh oh!

weiguangli-io commented Feb 15, 2026

Uh oh!

weiguangli-io commented Feb 17, 2026

Uh oh!

weiguangli-io commented Feb 17, 2026

Uh oh!

steipete commented Feb 27, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

weiguangli-io commented Feb 15, 2026 •

edited by greptile-apps bot

Loading