Add subscription-backed GPT lanes for Claude Code#3
Conversation
|
You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard. |
|
Warning Review limit reached
Next review available in: 53 minutes Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available. How can I continue?After more reviews become available, a review can be triggered using the To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews. How do review limits work?CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability. For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window. Please refer docs for additional details. Review details⚙️ Run configurationConfiguration used: defaults Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (3)
📝 WalkthroughWalkthroughAdds an opt-in ChangesClaude GPT integration
Estimated code review effort: 4 (Complex) | ~45 minutes Sequence Diagram(s)sequenceDiagram
participant User
participant claude-gpt
participant codex-auth
participant claude-code-proxy
participant ClaudeCode
User->>claude-gpt: Start GPT-backed Claude Code session
claude-gpt->>codex-auth: Export selected profile lease
claude-gpt->>claude-code-proxy: Start proxy and wait for healthz
claude-gpt->>ClaudeCode: Launch with mapped model lanes
claude-gpt->>codex-auth: Renew lease during session
ClaudeCode-->>claude-gpt: Return exit status
claude-gpt->>claude-code-proxy: Stop proxy and clean temporary state
🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 3
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@bin/claude-gpt`:
- Around line 448-452: Enforce the Bash 5.1+ requirement for the wait flow using
wait -p in the launcher, preferably through the script’s existing version-check
mechanism; otherwise document the minimum required Bash version prominently.
Ensure older shells fail with a clear actionable message before reaching the
wait -n invocation.
In `@lib/codex-auth/usage.sh`:
- Line 481: Update the coprocess write redirections in the affected initialize
request and its corresponding second site to explicitly redirect stdout with
1>&"$rate_in" before redirecting stderr to /dev/null. Preserve the existing
printf payload and control flow while replacing the ambiguous redirection order
at both sites.
In `@README.md`:
- Around line 342-343: Update the README cleanup command to use the same
`${PREFIX:-$HOME/.local}` installation directory as install.sh, ensuring all
listed binaries—including claude-gpt and claude-code-proxy—are removed from a
custom prefix.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro
Run ID: f79ac910-3309-4398-8124-9bedca05e3e5
⛔ Files ignored due to path filters (1)
tui/uv.lockis excluded by!**/*.lock
📒 Files selected for processing (10)
CHANGELOG.mdREADME.mdbin/claude-gptbin/codex-authinstall.shlib/codex-auth/claude-gpt.shlib/codex-auth/usage.shtests/run.shtui/pyproject.tomltui/src/codex_auth_tui/__init__.py
There was a problem hiding this comment.
♻️ Duplicate comments (1)
README.md (1)
381-382: 🎯 Functional Correctness | 🟡 MinorHonor the configured install prefix during cleanup.
The removal command hard-codes
~/.local/bin, while installation uses${PREFIX:-$HOME/.local}. Users with a customPREFIXcan therefore leaveclaude-gptandclaude-code-proxyinstalled. Use the same prefix expression asinstall.sh.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@README.md` around lines 381 - 382, Update the cleanup rm command in README.md to use the same ${PREFIX:-$HOME/.local}/bin prefix as install.sh instead of hard-coding ~/.local/bin, ensuring all listed binaries including claude-gpt and claude-code-proxy are removed for custom PREFIX values.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Duplicate comments:
In `@README.md`:
- Around line 381-382: Update the cleanup rm command in README.md to use the
same ${PREFIX:-$HOME/.local}/bin prefix as install.sh instead of hard-coding
~/.local/bin, ensuring all listed binaries including claude-gpt and
claude-code-proxy are removed for custom PREFIX values.
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro
Run ID: ce626050-cf93-41f9-bcf6-e7eac93cff7b
📒 Files selected for processing (5)
CHANGELOG.mdREADME.mdbin/claude-gptinstall.shtests/run.sh
🚧 Files skipped from review as they are similar to previous changes (3)
- install.sh
- tests/run.sh
- bin/claude-gpt
1 participant
What changed
claude-gpt, an opt-in Claude Code launcher backed by a saved ChatGPT/Codex subscription profile rather thanOPENAI_API_KEYcodex-authas the only refresh-token owner; the proxy receives a private, access-only lease snapshot with no refresh tokenultracodeworkflow orchestration without pinning a proxy-wide effort[1m]model hints while enforcing the real 372K Codex subscription ceilingcodex-authactive-profile changes through validated atomic lease swaps; explicit--profile NAMEsessions stay identity-pinnedclaude-code-proxy0.1.10-codex-auth.2The source-visible proxy fixes are also submitted upstream as raine/claude-code-proxy#28, #29, and #30.
Verification
0.1.10-codex-auth.2and the installed launcher byte-matches this branchPATCHED_CODEX_OKthrough GPT-5.6 SolCompatibility boundary
This remains a third-party Anthropic-to-Codex compatibility route; normal
claudestays unchanged. A proxy process already in memory keeps its current binary until that Claude session exits, while every newclaude-gptlaunch uses the installed.2build. Sol Ultra Fast consumes the selected profile's fast-mode quota.Summary by CodeRabbit
claude-gptlauncher to run Claude Code with GPT-backed model lanes, effort controls (includingultracode/ultra), and lane overrides.claude-gpt-exportto export ChatGPT-backed proxy authorization safely.claude-code-proxy(Linux) with optional opt-out.claude-gptand the proxy.