phuryn/product-engineering / ai-shipping/skills/independent-review
independent-review
VerifiedBefore 'ship it', a deploy or release, for 'an independent review', and after each fix until a round is clean: a fresh reviewer that never saw the build.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 29.8 KB |
| receipts.md | markdown | 15.6 KB |
| references/overlapping-phases.md | markdown | 3.7 KB |
| references/reviewer-clis.md | markdown | 4.7 KB |
| scripts/review.py | script | 68.3 KB |
references/reviewer-clis.md
View on GitHub# Reviewer CLI setup and diagnostics
Use this reference to check what a specific CLI isolates, prepare a manual round, or diagnose a failed one. The normal command and the evidence needed to accept a round are in `SKILL.md` section Running it.
## What each reviewer can still see
- A fresh Codex session in a read-only sandbox excludes user config and its MCP servers, execpolicy rules, connected apps, plugins, tool suggestions, web search, image generation, memory and saved sessions. A flag disables hooks. Repository and `CODEX_HOME` `AGENTS.md`, bundled skills, agent tools and administrator additions remain. Name them in the report (`codex-agents` § Reviewer points to the manual checks).
- A fresh headless Claude Code session uses read-only tools. It has no settings, hooks, plugins, MCP servers, auto-memory or saved session, starting from the repository and brief alone.
- Grok runs with read-only tools. Its memory, managed MCP servers, and Claude, Cursor and Codex compatibility, including session reading, are off for the run. The wrapper reads whatever still loads from `grok inspect` and prints it for the report. In a trusted folder, that includes project instructions and a gitignored `CLAUDE.local.md`. A review worktree of the snapshot does not hold that file.
A sandbox flag alone is insufficient. In Codex, `-s read-only` binds only shell commands; user configuration can add tools acting outside it (Codex CLI 0.159.0, observed `2026-10-29`). Ignoring user config does not remove connected apps, plugins or web search (Codex CLI 0.166.0, observed `2026-11-14`). Leave that configuration out and switch those features off by name; apply `codex-agents` § Reviewer.
## Wrapper flags and paging
**Use the repository's runnable fallback when the default reviewer cannot run.** `--reviewer auto` defaults to the first installed of Codex, Claude Code and Grok. Choose `--reviewer codex`, `claude` or `grok` explicitly. Codex runs `exec --json --ephemeral --ignore-user-config --ignore-rules -s read-only --disable apps --disable plugins --disable tool_suggest --disable image_generation --disable memories --disable hooks -c web_search=disabled`. It passes `--effort` as `model_reasoning_effort` (`xhigh` by default). Its event stream fails a round with no successful command, a completed outside MCP call, a web search, completed file changes, a missing report or events, or a failed turn. Pass the review model the user confirmed (`codex-agents`) with `--model`, because ignoring user config also omits your configured model. On Windows, `--windows-sandbox elevated|unelevated` explicitly overrides the mode; absent that flag, the wrapper restores `[windows].sandbox` from config. It stops with setup instructions when neither supplies a mode (observed `2026-11-16`, Codex CLI 0.169.0). Claude runs `claude -p` with `--restricted`, `--tools Read,Grep,Glob`, `--strict-mcp-config`, `--no-session-persistence`, `--effort max` and `CLAUDE_CODE_DISABLE_AUTO_MEMORY=1`. Measured on 2026-10-10: without that variable, a headless session loads the project's memory index holding builder notes. With these flags, no project instructions, personal notes, memory or skills list was in its context.
- `2026-11-16`, Codex CLI 0.169.0 on Windows: an elevated review timed out after 60 minutes with zero commands. The unelevated round completed in about 15 minutes. The wrapper therefore accepts an explicit Windows mode override; absence keeps config lookup.
Grok and Claude Code have no shell; Codex has a read-only shell. The wrapper writes the range's log and diffs, then everything uncommitted, to a never-committed file inside the git directory. It lists untracked files individually, with their content. It gives the reviewer the length and paging instructions: some file tools return at most 1,000 lines and may not say there are more. For Grok, it raises the turn cap when paging the subject needs more turns. The Codex and Claude Code runs have no turn cap; `--max-turns` does not apply to them.
## Fallback order
After Codex, the order is Claude Code then Grok. After Claude Code, it's Grok then Codex. After Grok, it's Codex then Claude Code. With no other installed, it prints the `--subject-only` route.
## Manual runs
**Apply the CLI's skill when running a reviewer by hand.** Apply `codex-agents` § Reviewer for the wrapper's Codex command and its manual checks. Apply `grok-agents` for Grok's bare call and each wrapper flag's purpose.
**The prompt comes from a file, the working directory is the repository under review, and stdin never waits on a terminal.** Close stdin with `< /dev/null` when the CLI takes the prompt through a file flag. When it reads stdin, as Codex does, pass the prompt file itself. Never leave stdin waiting on a terminal: that hangs without output. Long inline prompts trip permission classifiers.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deployment | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| xlsxanthropics/skillsUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .xltx, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| shipping-and-launchaddyosmani/agent-skillsPrepares production launches. Use when preparing to deploy to production, or when asking what needs to be in place before shipping. Use when you need a pre-launch checklist, when setting up monitoring, when planning a staged rollout, or when you need a rollback strategy. | Stars: 119k | Updated | Claude CodeCodex | Verified | Install |
| childhoodthedotmack/claude-memBabysit a shipped release after merge, publication, or deployment. Verify published packages, production behavior, installed plugins, and late CI or user reports until the observation window is clean; use babysit for PR readiness before release. | Stars: 118k | Updated | Claude CodeCodex | Verified | Install |