agentleFS
Sign inStart free
zirui-songzirui-song/claude-skills/skills/codex-review-plan/SKILL.md

codex-review-plan

Skill17 starsChanged 3 months agoMITView on GitHub

Use after drafting an implementation or research plan (in plan mode, or a ~/.claude/plans/ file) to get a second-opinion review from Codex against the actual codebase, then critically incorporate the feedback. Makes the Claude Code plan better via cross-model review. Trigger phrases include "send the plan to Codex", "have Codex review this plan", "get a second opinion on the plan".

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 1 is distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Compose the prompt: Use the *review template* from codex-bridge, with the plan text inside the <PLAN>…</PLAN> block.

    line 27
  2. Very specific:

    Summarize: Read $OUT (kept in the temp file).

    line 37
  3. Very specific:

    Revise + report: Update the plan file with the accepted changes.

    line 45
  4. Distinctive:

    If Codex returns nothing useful or errors, say so plainly.

    line 54

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

codex-review-plan (Claude → Codex)

Send a Claude-authored plan to Codex for an adversarial review against the real codebase, then critically fold in the good points. Use it right after drafting a plan and before committing to it.

Mechanics (wrapper, prompt template, timeout, output rules) live in the codex-bridge skill — read it first.

Workflow

  1. Resolve the plan. In priority order:
    • The active plan-mode plan file in ~/.claude/plans/ (the one for this session; if ambiguous, the most recently modified *.md).
    • Else a file path the user provided.
    • Else pasted plan text → write to a temp file.
  2. Resolve the target project dir (the codebase the plan is about). Infer from the plan's content (a referenced project path); else use cwd. --skip-git-repo-check handles non-repo research dirs.
  3. Compose the prompt. Use the review template from codex-bridge, with the plan text inside the <PLAN>…</PLAN> block. Write it to a temp prompt file.
  4. Call Codex via the wrapper, with a long Bash timeout (~600000 ms):
    OUT=/tmp/codex-plan-review-$$.md
bash ~/.claude/skills/codex-bridge/codex-exec.sh "$PROJECT_DIR" "$OUT" < "$PROMPT_FILE"
    For a quick pass, append -c model_reasoning_effort=high.
  5. Summarize. Read $OUT (kept in the temp file). Summarize the key points in ≤10 lines in chat. Don't paste the whole review.
  6. Incorporate critically. Invoke superpowers:receiving-code-review and follow it: evaluate each point on its merits, accept the valid ones, and push back — with reasons — on anything wrong or inapplicable. No performative agreement; verify against the code before changing the plan.
  7. Revise + report. Update the plan file with the accepted changes. Report, briefly: what changed, and what you rejected and why. Note the path to the full review ($OUT).

Notes

  • Plan mode: this is meant to run during plan mode. The codex exec Bash call will prompt for permission once; editing the plan file is already allowed.
  • Codex runs read-only — it cannot modify your files.
  • If Codex returns nothing useful or errors, say so plainly; don't fabricate a review.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free