From codex-advisor
Delegates implementation, fix, or investigation tasks to Codex via companion plugin, reviews results with git diffs, code evaluation, and saves reports. Use for 'codex rescue' requests.
How this skill is triggered — by the user, by Claude, or both
Slash command
/codex-advisor:codex-rescueThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Hand off a task to Codex via the Official companion's task subcommand. When Codex finishes, Claude reviews what was done.
Hand off a task to Codex via the Official companion's task subcommand. When Codex finishes, Claude reviews what was done.
CODEX_COMPANION=$("${CLAUDE_PLUGIN_ROOT}/scripts/resolve-companion.sh")
If resolve fails: direct to /codex-setup immediately. Do NOT proceed.
Record current state so we can diff after Codex runs:
git diff --stat
git stash list | head -1
Use the $CODEX_COMPANION resolved in Step 1:
node "$CODEX_COMPANION" task --write $ARGUMENTS
Pass through flags that companion supports: --background, --write, --model, --effort, --resume-last, --fresh.
Default to --write (Codex needs write access to implement fixes). If user explicitly asks for read-only investigation, omit --write.
Timeout: 300000ms (5 minutes) for foreground. Background tasks are tracked via /codex:status.
| Error | Action |
|---|---|
| "not authenticated" in stderr | Auth required → suggest codex login |
| Other error | Show raw error, don't retry silently |
Read ${CLAUDE_PLUGIN_ROOT}/references/evaluation.md.
git diff
git diff --stat
Review each changed file — read the actual code
Evaluate:
Report:
## Codex Implementation Review
### Changes Made
<git diff --stat>
### Evaluation
- [file]: [assessment]
### Agreement: <High|Partial|Disagreement>
### Verdict
<appropriate / has issues>
Apply standard Peer AI Evaluation from evaluation.md.
mkdir -p "${CLAUDE_PLUGIN_DATA}/reviews"
Save to ${CLAUDE_PLUGIN_DATA}/reviews/rescue-<YYYYMMDD-HHMMSS>.md.
/codex:status (Official command).npx claudepluginhub leejuoh/claude-code-zero --plugin codex-advisorDelegates coding tasks (debug, implement, refactor) to OpenAI Codex CLI via codex exec, skipping the Node companion runtime for faster execution. Codex writes code; Claude verifies.
Runs official Codex code review on code changes, diffs, branches, or commits, then double-checks each finding by reading source code and classifying agree/disagree/nuance.
Runs code review using OpenAI Codex CLI in parallel instances with fallback to Claude Code. Activates on 'review' commands or after implementation for quality checks.