codex-claude-bridge
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| OPENAI_API_KEY | No | Your OpenAI API key for pay-per-token usage. If not set, uses ChatGPT subscription via OAuth. | |
| RB_CONFIG_PATH | No | Explicit path to .reviewbridge.json config file. | |
| REVIEW_BRIDGE_DB | No | Path to SQLite database for persisting review history and session state. Defaults to 'reviews.db' in the current directory. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| review_planA | Get an independent code review of your implementation plan before writing code. Call this after drafting a plan and before implementing it. Returns a verdict (approve/revise/reject), findings, session_id, responding models, and persistence provenance. Pass the returned session_id to review_code later so the reviewer has full context. |
| review_codeA | Get an independent code review of your changes before committing. Call this after writing or modifying code. Pass a git diff as input. The diff parameter MUST contain actual git diff output (from git diff, gh pr diff, etc.), NOT a summary or description of changes. To review a branch or landed commits, pass base (and optionally head) instead and the bridge runs git diff base head in cwd. If you reviewed a plan first, pass the same session_id so the reviewer checks the code against the plan. Returns a verdict, findings, responding models, and persistence provenance. An auto-captured review also returns captured_from: the absolute directory the bridge ran git in. If that is not the repository you are working in, pass the diff explicitly. |
| review_precommitA | Final sanity check right before committing. Auto-captures staged git changes. Call this after git add and before git commit to catch last-minute issues. Returns ready_to_commit, blockers, warnings, responding models, and persistence provenance. An auto-captured check also returns captured_from: the absolute directory the bridge ran git in. If that is not the repository you are working in, pass the diff explicitly. |
| review_historyA | Look up past review results. Query by session_id to see all reviews in a session, or use last_n to get recent reviews. Results include immutable responding-model snapshots and a next_cursor for bounded pagination. A session_id query also returns the session's own state (in_progress / completed / failed with timestamps), so a review that failed or timed out is visible even though it produced no review row. |
| review_statusA | Check whether a review session is still running, completed, or failed. Use this if a review call timed out or you need to verify session state. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 5 tools
The tools are largely distinct: plan, code, precommit, history, and status each have a clear primary role. However, review_code and review_precommit both address reviewing changes before commit, and review_history also exposes session state that overlaps with review_status.
All tools follow a consistent review_<target> snake_case pattern, making their purpose predictable. The names are uniform and easy to group mentally.
Five tools is a well-scoped size for this domain. Each tool covers a distinct step in the review workflow without unnecessary bloat.
The set covers the full review lifecycle: plan review, code review, pre-commit sanity check, history lookup, and session status. Minor gaps include no explicit cancel/retry mechanism for failed or timed-out sessions, but the core workflow is complete.