review-by-opp:start
Start a guarded implementation + review session. Use when beginning work that should be reviewed by Codex before completion. Initializes the review ledger and configures the stop gate.
Best use case
review-by-opp:start is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Start a guarded implementation + review session. Use when beginning work that should be reviewed by Codex before completion. Initializes the review ledger and configures the stop gate.
Teams using review-by-opp:start should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/start/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How review-by-opp:start Compares
| Feature / Agent | review-by-opp:start | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Start a guarded implementation + review session. Use when beginning work that should be reviewed by Codex before completion. Initializes the review ledger and configures the stop gate.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
Related Guides
SKILL.md Source
You are starting a review-by-opp session. This activates strict review tracking.
## What to do
1. **Check prerequisites:**
- Verify this is a git repository. If not, tell the user: "review-by-opp requires a git repo. Run `git init` first."
- Check if a session already exists by looking for `reviews/current.json`. If one exists, ask: "A session is already active. Use `/review-by-opp:resume` to continue or `/review-by-opp:finalize` to close it first."
2. **Initialize session:**
- Create the `reviews/` directory structure: `reviews/current.json`, `reviews/rounds/`, `reviews/summaries/`
- Load config from `.review-by-opp.json` if it exists, otherwise use defaults
- Write initial `reviews/current.json` with:
```json
{
"session_id": "<uuid>",
"started_at": "<iso-timestamp>",
"updated_at": "<iso-timestamp>",
"current_round": 0,
"max_rounds": 4,
"findings": [],
"rounds": [],
"final_verdict": null,
"config": { ... }
}
```
3. **Check for Codex:**
- Run `npx @openai/codex --help` to verify Codex CLI is available
- If not installed, tell user: "Install Codex: `npm install -g @openai/codex` then `npx @openai/codex login`"
4. **Report status:**
- Show session ID, max rounds, review scope, and blocking severities
- Show Codex detection status
- Show Claude model/effort: "user-controlled (via Claude Code UI)"
- Show Codex config: model strategy, reasoning effort
5. **Set the rules:**
Tell the user:
> Review ledger session started. Rules:
> - After implementing changes, run `/review-by-opp:review` to get Codex feedback
> - Use `/review-by-opp:fix` to address findings
> - Use `/review-by-opp:status` to check progress
> - You cannot finalize until all blocking findings are resolved
> - Use `/review-by-opp:finalize` when ready to close
**CRITICAL:** You are the builder. Codex is the auditor. You must not claim work is complete while blocking findings remain open. The stop gate enforces this.
$ARGUMENTSRelated Skills
review-by-opp:status
Show current review ledger status including round count, open findings, blocking items, agent configuration, and whether the session can be finalized.
review-by-opp:review
Run Codex as an independent reviewer against the current diff or changed files. Parses findings into the ledger. Use after making code changes.
review-by-opp:resume
Resume an existing review-by-opp session. Use when returning to a previous review session.
review-by-opp:help
Learn about review-by-opp - what it does, how to use it, and available commands
review-by-opp:fix
Address Codex review findings from the ledger. Shows open findings and guides resolution. Each finding must get an explicit resolution state.
review-by-opp:finalize
Finalize the review session. Only succeeds if all blocking findings are resolved. Emits final verdict.
flutter-dart-code-review
库无关的Flutter/Dart代码审查清单,涵盖Widget最佳实践、状态管理模式(BLoC、Riverpod、Provider、GetX、MobX、Signals)、Dart惯用法、性能、可访问性、安全性和整洁架构。
security-review
Use this skill when adding authentication, handling user input, working with secrets, creating API endpoints, or implementing payment/sensitive features. Provides comprehensive security checklist and patterns.
addressing-pr-review-comments
Address all valid review comments on a PR for the current branch in the streamlit/streamlit repo. Covers both inline review comments and general PR (issue) comments. Use when a PR has reviewer feedback to address, including code changes, style fixes, and documentation updates.
lightning-architecture-review
Review Bitcoin Lightning Network protocol designs, compare channel factory approaches, and analyze Layer 2 scaling tradeoffs. Covers trust models, on-chain footprint, consensus requirements, HTLC/PTLC compatibility, liveness, and watchtower support.
gha-security-review
Find exploitable vulnerabilities in GitHub Actions workflows. Every finding MUST include a concrete exploitation scenario — if you can't build the attack, don't report it.
gh-review-requests
Fetch unread GitHub notifications for open PRs where review is requested from a specified team or opened by a team member. Use when asked to "find PRs I need to review", "show my review requests", "what needs my review", "fetch GitHub review requests", or "check team review queue".