docs(providers/openai): note OpenAI Realtime requires Platform credits, not Codex/ChatGPT subscription by lonexreb · Pull Request #77215 · openclaw/openclaw

lonexreb · 2026-05-04T08:51:37Z

Summary

Adds an inline clarification to docs/providers/openai.md that OpenAI Realtime voice (used by Voice Call realtime.provider: "openai" and Control UI Talk with talk.realtime.provider: "openai") is billed against OpenAI Platform credits rather than Codex/ChatGPT subscription quota. Documents both supported auth routes — direct Platform OPENAI_API_KEY and the openai-codex OAuth fallback that mints Realtime client secrets through the Platform API.

Why

From #76498: a user with healthy Codex OAuth (where openai-codex/* chat worked fine) hit You exceeded your current quota on the first Realtime turn, even though the ChatGPT/Codex usage dashboard still showed remaining quota. The Realtime route goes through the public OpenAI Platform Realtime API, so the underlying organization needs funded Platform credits — independent of which auth path mints the client secret. Nothing in the docs surfaced that distinction, so troubleshooting was guesswork.

Changes

docs/providers/openai.md
- Append a qualifier to the Realtime voice row of the OpenClaw feature coverage matrix: Yes (requires OpenAI Platform credits, not Codex/ChatGPT subscription), with the surface column updated to include the correct talk.realtime.provider: "openai" Control UI Talk key.
- Add a <Note> block under the matrix that:
  - Names the affected surfaces with current config keys (realtime.provider: "openai" for Voice Call, talk.realtime.provider: "openai" for Control UI Talk)
  - Explains why Codex/ChatGPT subscription quota does not apply (different billing path)
  - Names the exact symptom (insufficient_quota / "You exceeded your current quota" on the first Realtime turn even when chat works)
  - Documents both supported auth routes: a Platform OPENAI_API_KEY (via talk.realtime.providers.openai.apiKey or plugins.entries.voice-call.config.realtime.providers.openai.apiKey) or an openai-codex OAuth profile whose underlying organization has Platform billing — both routes mint Realtime client secrets through the Platform API, so either way the org needs funded Platform credits
  - Gives the concrete fix (top up Platform credits at platform.openai.com/account/billing)
CHANGELOG.md — Unreleased > Changes entry.

Test plan

pnpm exec oxfmt --check --threads=1 docs/providers/openai.md CHANGELOG.md — clean
git diff --check — clean
No code changes; docs-only.
Mintlify-compatible: kept the <Note> JSX block, no em-dashes inside headings.
Rebased onto current origin/main (HEAD 8dc213227b); preserves the codex-cli/<model> legacy-row addition that landed on main.

Closes #76498.

Real behavior proof

Behavior or issue addressed: Fixes Docs/UX: Realtime Talk with OpenAI requires OpenAI Platform API credits, not Codex/ChatGPT subscription quota #76498. Users with healthy openai-codex OAuth who tried OpenAI Realtime voice got insufficient_quota on the first Realtime turn even though their Codex subscription worked fine for chat. The Realtime route is billed against OpenAI Platform credits regardless of whether the credential is a direct Platform API key or an openai-codex OAuth profile — both mint client secrets through the Platform API. The original docs didn't mention this; users had no signposting from "Codex works" to "but Realtime needs Platform billing on the same org."
Real environment tested: The patched docs/providers/openai.md as rendered by Mintlify (the production docs hosting layer at docs.openclaw.ai). Config keys verified against current main:
- talk.realtime.provider / talk.realtime.providers.openai.apiKey — src/gateway/server-methods/talk-shared.ts:112 (Talk realtime config reads config.talk?.realtime, not the top-level Talk speech provider map).
- openai-codex OAuth fallback — extensions/openai/realtime-voice-provider.ts:304 calls resolveProviderAuthProfileApiKey({ provider: "openai-codex" }) when no direct key is present; native bridge at extensions/openai/realtime-voice-provider.ts:645 falls through to OAuth-backed client-secret creation.
- Test coverage — extensions/openai/realtime-voice-provider.test.ts:282 covers minting an ephemeral Realtime secret from an openai-codex OAuth token when no direct API key is configured.

Exact steps or command run after this patch (HEAD 768259819e, rebased onto origin/main at 8dc213227b):

$ pnpm exec oxfmt --check --threads=1 docs/providers/openai.md CHANGELOG.md
Checking formatting...

All matched files use the correct format.

$ git diff --check
(clean)

$ git diff --stat origin/main..HEAD
 CHANGELOG.md             |  1 +
 docs/providers/openai.md | 50 ++++++++++++++++++++++++++++++++++--------------
 2 files changed, 37 insertions(+), 14 deletions(-)

Evidence after fix (current HEAD 768259819e): The rendered <Note> block at docs/providers/openai.md:99 now contains the corrected, OAuth-aware remediation:

OpenAI Realtime voice (used by Voice Call's realtime.provider: "openai" and Control UI Talk with talk.realtime.provider: "openai") goes through the public OpenAI Platform Realtime API, which is billed against OpenAI Platform credits rather than Codex/ChatGPT subscription quota. An account with healthy Codex OAuth that runs openai-codex/* chat models without issue can still hit insufficient_quota / "You exceeded your current quota" on the first Realtime turn if the same OpenAI organization has no Platform billing set up.

Fix: top up Platform credits at platform.openai.com/account/billing for the organization backing your realtime credentials. Realtime accepts either a Platform OPENAI_API_KEY (configured via talk.realtime.providers.openai.apiKey for Control UI Talk, or plugins.entries.voice-call.config.realtime.providers.openai.apiKey for Voice Call) or an openai-codex OAuth profile whose underlying organization has Platform billing — both routes mint Realtime client secrets through the Platform API, so either way the org needs funded Platform credits. For chat turns you can still use openai-codex/* against the same OpenClaw install; Realtime is the one route that needs Platform billing.

The Realtime voice row in the capability table at docs/providers/openai.md:96 carries the updated Yes (requires OpenAI Platform credits, not Codex/ChatGPT subscription) status note and includes talk.realtime.provider: "openai" for Control UI Talk in the surface column.
Observed result after fix: A user hitting insufficient_quota on their first Realtime turn while their openai-codex/* chat keeps working can now find the explanation by reading the OpenAI provider docs — the capability table flags the Platform-credits requirement on the Realtime row, and the <Note> block tells them both (a) which org's billing they need to top up and (b) which config keys to point at the funded credential, including the OAuth path. Pre-fix, the docs page did not mention the billing split at all and operators had no signposting from "Codex works" to "but Realtime needs Platform billing."
What was not tested: Live insufficient_quota reproduction against a real OpenAI organization without Platform billing — the bug is a documentation gap, not a code regression; the underlying API contract (Codex subscription ≠ Platform credits) is established by OpenAI's billing system and verified by the original reporter in Docs/UX: Realtime Talk with OpenAI requires OpenAI Platform API credits, not Codex/ChatGPT subscription quota #76498.

chatgpt-codex-connector

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 725a5371a9

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

Open a pull request for review
Mark a draft as ready
Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector · 2026-05-04T08:54:08Z

+Fix: top up Platform credits at
+[platform.openai.com/account/billing](https://platform.openai.com/account/billing)
+and configure `OPENAI_API_KEY` (or `talk.providers.openai.apiKey` /
+`voice.realtime.openai.apiKey`) with a key from that organization. For chat


Replace nonexistent voice config path

Update this path because voice.realtime.openai.apiKey is not a recognized config key, so users who follow the note will set a value that OpenClaw never reads. The realtime voice provider consumes plugins.entries.voice-call.config.realtime.providers.openai.apiKey (or OPENAI_API_KEY), so the current doc can leave Voice Call realtime misconfigured even after applying the suggested fix.

Useful? React with 👍 / 👎.

clawsweeper · 2026-05-04T08:54:51Z

Codex review: needs changes before merge. Reviewed May 27, 2026, 2:41 AM ET / 06:41 UTC.

Summary
The branch adds an OpenAI provider docs note explaining that Realtime voice uses OpenAI Platform billing/credits and documents the direct API-key and openai-codex OAuth credential paths, plus a changelog entry.

PR surface: Docs +23. Total +23 across 2 files.

Reproducibility: not applicable. as a docs PR; the linked report supplies the user confusion and current source confirms the documented config/auth paths.

Review metrics: 1 noteworthy metric.

Release-owned changelog edit: 1 added entry. Root policy reserves CHANGELOG.md for release generation, so maintainers should notice this before merge.

Merge readiness
Overall: 🦐 gold shrimp
Proof: 🐚 platinum hermit
Patch quality: 🦐 gold shrimp
Result: needs maintainer review before merge.

Overall follows the weaker of proof and patch quality, so missing proof can cap an otherwise strong patch.

Rank-up moves:

Remove the added CHANGELOG.md line.
Refresh the PR proof text to name the current head SHA if maintainers want exact-head proof after the rebase.

Risk before merge

Merging the PR as-is would put a normal contributor docs entry into release-owned CHANGELOG.md, creating release-generation/process churn that green CI does not settle.

Maintainer options:

Drop the changelog edit before merge (recommended)
Remove the added CHANGELOG.md line and keep the release-note context in the PR body or commit message so release generation remains owner-controlled.
Accept release-process churn
Maintainers could intentionally keep the changelog entry, but that would bypass the repository's release-owned changelog flow for a normal docs PR.

Copy recommended automerge instruction

@clawsweeper automerge

Special instructions:
Remove only the added `CHANGELOG.md` entry from this PR; do not change the `docs/providers/openai.md` clarification, and keep release-note context in the PR body/commit message.

Next step before merge
A narrow automated repair can remove the release-owned changelog entry; maintainer review can then decide whether the docs wording is ready to land.

Security
Cleared: The diff only changes documentation/changelog text and does not introduce code execution, dependency, credential, workflow, or supply-chain risk.

Review findings

[P2] Remove the release-owned changelog entry — CHANGELOG.md:1854

Review details

Best possible solution:

Keep the OpenAI provider docs clarification, remove the CHANGELOG.md edit, and let release-note context live in the PR body or final squash message.

Do we have a high-confidence way to reproduce the issue?

Not applicable as a docs PR; the linked report supplies the user confusion and current source confirms the documented config/auth paths.

Is this the best way to solve the issue?

No as submitted: the docs note is the right narrow fix, but the normal PR should not carry a CHANGELOG.md edit.

Full review comments:

[P2] Remove the release-owned changelog entry — CHANGELOG.md:1854
CHANGELOG.md is release-owned in this repo, and normal PRs should keep release-note context in the PR body or commit message instead. Please drop this added line and leave the OpenAI provider docs change as the PR content.
Confidence: 0.93

Overall correctness: patch is incorrect
Overall confidence: 0.9

AGENTS.md: found and applied where relevant.

Codex review notes: model gpt-5.5, reasoning high; reviewed against 8c8162f1f7d7.

Label changes

Label changes:

add merge-risk: 🚨 automation: The PR edits release-owned changelog automation output, which can create release-process drift even when normal checks pass.
add proof: sufficient: Contributor real behavior proof is sufficient. The PR body supplies terminal formatting/diff-check output and the after-fix rendered docs text for this docs/changelog-only clarification, though the proof text still names an earlier head SHA after a rebase.
add rating: 🦐 gold shrimp: Overall readiness is 🦐 gold shrimp; proof is 🐚 platinum hermit and patch quality is 🦐 gold shrimp.
add status: ⏳ waiting on author: ClawSweeper has contributor-facing work open and is waiting for author action. Sufficient (terminal): The PR body supplies terminal formatting/diff-check output and the after-fix rendered docs text for this docs/changelog-only clarification, though the proof text still names an earlier head SHA after a rebase.
remove rating: 🐚 platinum hermit: Current PR rating is rating: 🦐 gold shrimp, so this older rating label is no longer current.
remove status: 👀 ready for maintainer look: Current PR status label is status: ⏳ waiting on author.

Label justifications:

P3: This is a small docs/process cleanup for a useful clarification, with no runtime regression in the patch.
merge-risk: 🚨 automation: The PR edits release-owned changelog automation output, which can create release-process drift even when normal checks pass.
rating: 🦐 gold shrimp: Overall readiness is 🦐 gold shrimp; proof is 🐚 platinum hermit and patch quality is 🦐 gold shrimp.
status: ⏳ waiting on author: ClawSweeper has contributor-facing work open and is waiting for author action. Sufficient (terminal): The PR body supplies terminal formatting/diff-check output and the after-fix rendered docs text for this docs/changelog-only clarification, though the proof text still names an earlier head SHA after a rebase.
proof: sufficient: Contributor real behavior proof is sufficient. The PR body supplies terminal formatting/diff-check output and the after-fix rendered docs text for this docs/changelog-only clarification, though the proof text still names an earlier head SHA after a rebase.

Evidence reviewed

PR surface:

Docs +23. Total +23 across 2 files.

View PR surface stats

Area	Files	Added	Removed	Net
Source	0	0	0	0
Tests	0	0	0	0
Docs	2	37	14	+23
Config	0	0	0	0
Generated	0	0	0	0
Other	0	0	0	0
Total	2	37	14	+23

Acceptance criteria:

git diff --check
pnpm exec oxfmt --check --threads=1 docs/providers/openai.md

What I checked:

Root policy blocks normal PR changelog edits: Root policy says CHANGELOG.md is release-owned and normal PRs should put release-note context in the PR body, squash message, or direct commit instead. (AGENTS.md:191, 8c8162f1f7d7)
PR adds one release-owned changelog entry: The PR adds a single CHANGELOG.md entry for the OpenAI Realtime docs clarification, which is the concrete merge blocker. (CHANGELOG.md:1854, 1c309e669519)
Docs change targets a real gap in the feature matrix: Current main lists Realtime voice as available but does not mention the Platform-credit billing distinction in the OpenAI feature coverage table. Public docs: docs/providers/openai.md. (docs/providers/openai.md:96, 8c8162f1f7d7)
PR docs text uses the current Talk realtime config surface: The PR's table and note use talk.realtime.provider and talk.realtime.providers.openai.apiKey, matching the current Talk realtime config reader. Public docs: docs/providers/openai.md. (docs/providers/openai.md:96, 1c309e669519)
Source confirms Talk reads config.talk.realtime: buildTalkRealtimeConfig reads config.talk?.realtime, uses talkRealtime.provider, and merges talkRealtime.providers over Voice Call realtime provider config. (src/gateway/server-methods/talk-shared.ts:113, 8c8162f1f7d7)
Source confirms Voice Call API-key path and OAuth fallback: The OpenAI realtime voice provider normalizes plugins.entries.voice-call.config.realtime.providers.openai.apiKey, falls back to openai-codex OAuth for GPT realtime models, and mints Realtime client secrets through the OpenAI Platform endpoint. (extensions/openai/realtime-voice-provider.ts:202, 8c8162f1f7d7)

Likely related people:

vincentkoc: Current blame and recent git history show Vincent Koc on the OpenAI provider docs table and adjacent docs/config surface. (role: recent area contributor; confidence: medium; commits: fca77dcb19c8, 90fac509875e, 1cff54c783a1; files: docs/providers/openai.md, src/gateway/server-methods/talk-shared.ts, extensions/openai/realtime-voice-provider.ts)
steipete: Recent history shows Peter Steinberger carrying OpenAI realtime provider and Talk config refactors, and the PR discussion explicitly flagged @steipete as recent area owner for review. (role: realtime provider/config contributor; confidence: medium; commits: a5b563280910, 6dfdc92bd4c6, a23ab9b906dc; files: extensions/openai/realtime-voice-provider.ts, docs/providers/openai.md, src/config/talk.ts)

What the crustacean ranks mean

🦀 challenger crab: rare, exceptional readiness with strong proof, clean implementation, and convincing validation.
🦞 diamond lobster: very strong readiness with only minor maintainer review expected.
🐚 platinum hermit: good normal PR, likely mergeable with ordinary maintainer review.
🦐 gold shrimp: useful signal, but proof or patch confidence is still limited.
🦪 silver shellfish: thin signal; proof, validation, or implementation needs work.
🧂 unranked krab: not merge-ready because proof is missing/unusable or there are serious correctness or safety concerns.
🌊 off-meta tidepool: rating does not apply to this item.

Shiny media proof means a screenshot, video, or linked artifact directly shows the changed behavior. Runtime, network, CSP, and security claims still need visible diagnostics.

How this review workflow works

ClawSweeper keeps one durable marker-backed review comment per issue or PR.
Re-runs edit this comment so the latest verdict, findings, and automation markers stay together instead of adding duplicate bot comments.
A fresh review can be triggered by eligible @clawsweeper re-review comments, exact-item GitHub events, scheduled/background review runs, or manual workflow dispatch.
PR/issue authors and users with repository write access can comment @clawsweeper re-review or @clawsweeper re-run on an open PR or issue to request a fresh review only.
Maintainers can also comment @clawsweeper review to request a fresh review only.
Fresh-review commands do not start repair, autofix, rebase, CI repair, or automerge.
Maintainer-only repair and merge flows require explicit commands such as @clawsweeper autofix, @clawsweeper automerge, @clawsweeper fix ci, or @clawsweeper address review.
Maintainers can comment @clawsweeper explain to ask for more context, or @clawsweeper stop to stop active automation.

lonexreb · 2026-05-04T19:23:28Z

Addressed [P2] from codex review (docs/providers/openai.md:105). The current PR head already replaces voice.realtime.openai.apiKey with the actual config keys: talk.providers.openai.apiKey for Control UI Talk and plugins.entries.voice-call.config.realtime.providers.openai.apiKey for Voice Call. Codex was reviewing an older revision (725a5371a9); the current head (f954d6cd97 after rebase) carries the fix. Also rebased onto current origin/main.

lonexreb · 2026-05-12T17:04:25Z

Requesting proof: override label from a maintainer.

This is a docs-only PR that adds an inline note to docs/providers/openai.md warning that OpenAI Realtime voice requires OpenAI Platform credits (not Codex/ChatGPT subscription quota). There is no code change to exercise, and the policy-required "Real behavior proof" evidence categories (terminal output, runtime logs, etc.) don't have a clean fit for a docs-only diff.

The supporting context for the documented behavior (Codex OAuth account hitting insufficient_quota on Realtime turns when no Platform billing is configured) is the user-reported pattern in #76498. The corrected config path (plugins.entries.voice-call.config.realtime.providers.openai.apiKey for Voice Call, talk.providers.openai.apiKey for Control UI Talk) is the one the bundled openai plugin actually consumes — codex round-1 review caught a voice.realtime.openai.apiKey typo and the current HEAD has it fixed.

Happy to add proof: supplied content if a maintainer wants me to translate the documented failure mode into a synthetic-but-real-environment proof from my install, but providing screenshots of someone else's billing console / live Realtime session feels like the wrong kind of evidence for a docs change. Please apply proof: override or advise.

…lback Address Codex review on PR openclaw#77215: - Use talk.realtime.provider / talk.realtime.providers.openai.apiKey instead of the speech/TTS talk.provider keys - Stop implying API-key auth is the only Realtime route; document the openai-codex OAuth fallback that mints Realtime client secrets Realtime still requires Platform credits, but the credentials path can be either a Platform OPENAI_API_KEY or an openai-codex OAuth profile whose organization has funded Platform credits.

lonexreb · 2026-05-17T10:07:11Z

@steipete — flagging you for review since codex identified you as the recent area owner of the OpenAI realtime provider and docs/providers/openai.md. This is a small docs-only clarification noting that OpenAI Realtime voice is billed against Platform credits (covers both Platform OPENAI_API_KEY and the openai-codex OAuth fallback). Codex cleared the patch; just needs maintainer eyes. Thanks.

…s, not Codex/ChatGPT subscription (openclaw#76498) Realtime voice (Voice Call realtime.provider: openai and Control UI Talk with talk.provider: openai) goes through the public OpenAI Platform Realtime API and needs an OPENAI_API_KEY whose organization has funded Platform credits. Codex/ChatGPT subscription quota does not cover this route, so an install whose openai-codex/* chat works fine can still hit insufficient_quota / 'You exceeded your current quota' on the first Realtime turn. Adds an inline note next to the Realtime row of the OpenAI capability matrix and a follow-up Note block that explains the symptom, the fix (top up Platform credits + configure OPENAI_API_KEY), and why openai-codex is not registered as a realtime voice provider. This was the central confusion in openclaw#76498. Documenting it inline prevents the same fail-then-discover loop for the next user.

clawsweeper · 2026-05-27T06:41:24Z

ClawSweeper PR egg

🔥 Warming up: real-behavior proof passed; findings, security review, or rank-up moves are still in progress.

Hatch command

Comment @clawsweeper hatch when this PR is hatchable.

Hatchability rules:

Merged PRs are hatchable.
Open PRs are hatchable when they are status: 👀 ready for maintainer look, status: 🚀 automerge armed, or labeled clawsweeper:automerge.
Closed unmerged PRs are hatchable only when one of those hatchable labels is still present in the durable record.

What is this egg doing here?

Eggs appear after the PR passes real-behavior proof. It is here for vibes, not verdicts: it does not change labels, ratings, merge decisions, or automation.
The shell reacts to review momentum: open follow-up work warms it up, re-review makes it wobble, and a clean final review lets it hatch.
Hatchability usually comes from sufficient real-behavior proof, no blocking P0/P1/P2 findings, no security attention needed, and clean correctness. A merged PR is already final, so merge makes the egg hatchable independently.
The hatch is seeded from this repository and PR number, so the same PR keeps the same creature; the reviewed head SHA can only change safe visual details.
Rarity is just collectible sparkle: 🥚 common, 🌱 uncommon, 💎 rare, ✨ glimmer, and 🌈 legendary.

steipete · 2026-05-27T07:51:14Z

Maintainer verification before merge:

Exact head: 1c309e669519b037352e0947c6a693df57bf14cf
Local commands run:
- pnpm docs:list
- git diff --check origin/main...refs/remotes/pr/77215
- source checks for talk.realtime.provider, talk.realtime.providers.openai.apiKey, plugins.entries.voice-call.config.realtime.providers.openai.apiKey, and the openai-codex OAuth Realtime client-secret path
CI/checks green on current head:
- check-docs success, 2026-05-27T06:32:05Z
- preflight success, 2026-05-27T06:31:15Z
- security-fast success, 2026-05-27T06:31:22Z
- Real behavior proof success, 2026-05-27T06:46:40Z
- Socket PR/project checks success, 2026-05-27T07:23Z
Known gap: no live unfunded-OpenAI-org Realtime repro; accepted for this docs-only clarification because the linked issue supplies the observed failure and current source confirms the documented config/auth paths.

Maintainer note: changelog entry accepted for this landing.

openclaw-barnacle Bot added docs Improvements or additions to documentation size: XS labels May 4, 2026

chatgpt-codex-connector Bot reviewed May 4, 2026

View reviewed changes

lonexreb force-pushed the docs/76498-realtime-platform-credits branch 2 times, most recently from 08837fc to f954d6c Compare May 4, 2026 19:23

lonexreb force-pushed the docs/76498-realtime-platform-credits branch from f954d6c to cb74204 Compare May 5, 2026 18:12

openclaw-barnacle Bot added the triage: needs-real-behavior-proof Candidate: external PR needs after-fix proof from a real setup. label May 5, 2026

lonexreb force-pushed the docs/76498-realtime-platform-credits branch from cb74204 to de3da6c Compare May 12, 2026 23:25

openclaw-barnacle Bot added proof: supplied External PR includes structured after-fix real behavior proof. and removed triage: needs-real-behavior-proof Candidate: external PR needs after-fix proof from a real setup. labels May 12, 2026

clawsweeper Bot added the proof: sufficient ClawSweeper judged the real behavior proof convincing. label May 12, 2026

This was referenced May 14, 2026

Docs/UX: Realtime Talk with OpenAI requires OpenAI Platform API credits, not Codex/ChatGPT subscription quota #76498

Closed

Docs/Feature Request: Explain Realtime Talk voices, voice-agent role, and mobile/phone bridge options #76952

Closed

openclaw-barnacle Bot removed the proof: sufficient ClawSweeper judged the real behavior proof convincing. label May 17, 2026

clawsweeper Bot added P3 Low-priority cleanup, docs, polish, ergonomics, or speculative work. impact:auth-provider Auth, provider routing, model choice, or SecretRef resolution may break. labels May 17, 2026

lonexreb force-pushed the docs/76498-realtime-platform-credits branch from 775197f to 7682598 Compare May 17, 2026 06:45

openclaw-barnacle Bot added triage: needs-real-behavior-proof Candidate: external PR needs after-fix proof from a real setup. and removed proof: supplied External PR includes structured after-fix real behavior proof. labels May 17, 2026

clawsweeper Bot added the proof: sufficient ClawSweeper judged the real behavior proof convincing. label May 17, 2026

clawsweeper Bot added proof: sufficient ClawSweeper judged the real behavior proof convincing. rating: 🐚 platinum hermit Good normal PR readiness with ordinary maintainer review expected. labels May 19, 2026

clawsweeper Bot added status: 👀 ready for maintainer look ClawSweeper has no concrete contributor-facing blocker left for this PR. and removed impact:auth-provider Auth, provider routing, model choice, or SecretRef resolution may break. labels May 19, 2026

lonexreb force-pushed the docs/76498-realtime-platform-credits branch from 7682598 to 1c309e6 Compare May 27, 2026 06:30

openclaw-barnacle Bot removed the proof: sufficient ClawSweeper judged the real behavior proof convincing. label May 27, 2026

steipete merged commit 717003a into openclaw:main May 27, 2026
63 of 71 checks passed

sergiopesch mentioned this pull request Jun 4, 2026

[Bug]: OpenAI OAuth Realtime Talk sessions fail /v1/realtime/calls with HTTP 500 #90456

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

docs(providers/openai): note OpenAI Realtime requires Platform credits, not Codex/ChatGPT subscription#77215

docs(providers/openai): note OpenAI Realtime requires Platform credits, not Codex/ChatGPT subscription#77215
steipete merged 1 commit into
openclaw:mainfrom
lonexreb:docs/76498-realtime-platform-credits

lonexreb commented May 4, 2026 •

edited

Loading

Uh oh!

chatgpt-codex-connector Bot left a comment

Uh oh!

chatgpt-codex-connector Bot May 4, 2026

Uh oh!

clawsweeper Bot commented May 4, 2026 •

edited

Loading

Uh oh!

lonexreb commented May 4, 2026

Uh oh!

lonexreb commented May 12, 2026

Uh oh!

lonexreb commented May 17, 2026

Uh oh!

clawsweeper Bot commented May 27, 2026

Uh oh!

steipete commented May 27, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Uh oh!

Conversation

lonexreb commented May 4, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Why

Changes

Test plan

Real behavior proof

Uh oh!

chatgpt-codex-connector Bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

chatgpt-codex-connector Bot May 4, 2026

Choose a reason for hiding this comment

Uh oh!

clawsweeper Bot commented May 4, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

lonexreb commented May 4, 2026

Uh oh!

lonexreb commented May 12, 2026

Uh oh!

lonexreb commented May 17, 2026

Uh oh!

clawsweeper Bot commented May 27, 2026

Hatch command

Uh oh!

steipete commented May 27, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

lonexreb commented May 4, 2026 •

edited

Loading

clawsweeper Bot commented May 4, 2026 •

edited

Loading