codex-chef-workflows — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited codex-chef-workflows (Plugin) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
<p align="center"> <img src="assets/icon.svg" alt="Codex Chef icon" width="120" /> <br /> <img src="assets/banner.svg" alt="Codex Chef banner showing agents, MCPs, skills, verification, and multilingual docs" width="100%" /> </p>
<p align="center"> <a href="https://github.com/ucsahinn/codex-chef/actions/workflows/validate.yml"><img alt="Validate workflow" src="https://github.com/ucsahinn/codex-chef/actions/workflows/validate.yml/badge.svg" /></a> <a href="LICENSE"><img alt="MIT license" src="https://img.shields.io/github/license/ucsahinn/codex-chef?color=0f766e" /></a> <a href="README.md"><img alt="Documentation languages" src="https://img.shields.io/badge/docs-6%20languages-0f766e" /></a> <img alt="Windows and WSL ready" src="https://img.shields.io/badge/platform-Windows%20%2B%20WSL-164e63" /> </p>
<p align="center"> <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f310.svg" alt="" aria-hidden="true" width="20"> <strong>Docs:</strong> <a href="README.de.md"><img src="https://flagcdn.com/w20/de.png" alt="Deutsch" width="20"></a> | <a href="README.es.md"><img src="https://flagcdn.com/w20/es.png" alt="Español" width="20"></a> | <a href="README.md"><img src="https://flagcdn.com/w20/gb.png" alt="English" width="20"></a> | <a href="README.pt-BR.md"><img src="https://flagcdn.com/w20/br.png" alt="Português (Brasil)" width="20"></a> | <a href="README.tr.md"><img src="https://flagcdn.com/w20/tr.png" alt="Türkçe" width="20"></a> | <a href="README.fr.md"><img src="https://flagcdn.com/w20/fr.png" alt="Français" width="20"></a> </p>
<!-- bilingual-welcome:start --> <table> <tr> <td width="50%" valign="top"> <h3><img src="https://flagcdn.com/w20/gb.png" alt="English" width="20"> English welcome</h3> <p><strong>Codex Chef is a Windows-first Codex setup kit.</strong> It installs a reviewable baseline for agents, MCPs, skills, plugins, safety rules, and verification without turning on risky account or database access by default.</p> <p><strong>Start here:</strong> run the safe preview, then use the interactive installer only after you see what will change.</p> </td> <td width="50%" valign="top"> <h3><img src="https://flagcdn.com/w20/tr.png" alt="Türkçe" width="20"> Türkçe karşılama</h3> <p><strong>Codex Chef, Windows odaklı bir Codex kurulum kitidir.</strong> Ajanlar, MCP'ler, skill'ler, plugin akışları, güvenlik kuralları ve doğrulamayı tek yerde toplar; riskli hesap/database bağlantılarını kendiliğinden açmaz.</p> <p><strong>Buradan başla:</strong> önce safe preview çalıştır, sonra değişiklikleri gördükten sonra interactive install kullan.</p> </td> </tr> </table> <!-- bilingual-welcome:end -->
This is an unofficial community starter, not an OpenAI product. It is mapped to current official Codex documentation and keeps destructive, credentialed, publishing, deployment, database, and broad filesystem actions approval-gated.
The multilingual README entry points and six-language deep docs coverage are part of the release surface. English and Turkish keep the most complete operator flow; Deutsch, Español, Português (Brasil), and Français provide concise local entry points and links into the same deep docs.
Preview first:
git clone https://github.com/ucsahinn/codex-chef.git
cd codex-chef
powershell.exe -NoProfile -ExecutionPolicy Bypass -File .\scripts\install.ps1 -All -WhatIf
node scripts/plan-install.mjs --all --json --redact-pathsInstall after reviewing the preview:
Set-ExecutionPolicy -Scope Process Bypass -Force
.\scripts\install.ps1 -All -InteractiveBash or WSL:
git clone https://github.com/ucsahinn/codex-chef.git
cd codex-chef
chmod +x scripts/install.sh
./scripts/install.sh --all --interactiveRepair an existing Codex Chef setup without deleting user skills:
powershell.exe -NoProfile -ExecutionPolicy Bypass -File .\scripts\install.ps1 -Repair -WhatIf
powershell.exe -NoProfile -ExecutionPolicy Bypass -File .\scripts\install.ps1 -RepairUse npm run chef for the menu. Keep these commands handy:
| Need | Command |
|---|---|
| Fast local health check | npm run chef -- --status --repo-only --no-log |
| Full status board | npm run codex:status |
| Install preview | npm run chef -- --preview |
| Routing profile map | npm run chef -- --routing --profile starter-health |
| Diagnostics hub | npm run chef -- --diagnostics --no-log |
| Process audit only | npm run chef -- --processes --no-log |
| Repair preview/apply | npm run chef -- --repair then npm run chef -- --repair --apply |
Rules of thumb:
--status, --preview, --skills, --mcp, --routing, --diagnostics, --processes, --auth, and --logs are inspection-first surfaces.--apply or typed confirmation. The menu labels the write boundary before doing anything risky.--no-log when you want a strict no-log audit. Normal CLI logs are repo-local, ignored, and redacted.npx run; use npm run chef or npm --prefix <repo> run chef.Detailed CLI behavior lives in Install, Expected output, and Troubleshooting.
Codex Chef installs a reviewed Codex baseline, not a hidden copy of someone else's machine.
| Surface | What lands on your machine |
|---|---|
| <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f916.svg" alt="" aria-hidden="true" width="20"> Agent team | 21 Codex subagent role files under ~/.codex/agents/*.toml, including readable nickname_candidates. They are role definitions, not always-on services. |
| <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f9e0.svg" alt="" aria-hidden="true" width="20"> Durable guidance | Global ~/.codex/AGENTS.md with routing, verification, safety, and approval rules. |
| <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f50c.svg" alt="" aria-hidden="true" width="20"> MCP defaults | 7 useful MCP entries enabled for docs, code navigation, browser evidence, reasoning, and non-secret memory; 8 account/database/high-risk connectors disabled. |
| <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f9e9.svg" alt="" aria-hidden="true" width="20"> Plugin + skills | Local codex-chef-workflows plugin, three bundled skills, and sixteen reviewed optional global skills. |
| <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f6e1.svg" alt="" aria-hidden="true" width="20"> Safety gates | Dry runs, backups, validation, secret scanning, and approval gates before risky actions. |
Installed skills do not execute by themselves. They enter context when named by the user or when a task clearly matches their description. Codex subagents also do not spawn from every prompt; current Codex releases require the user to explicitly ask for subagents or parallel agent work.
These are the visible specialist names Codex Chef installs. They are role files for explicit, visible delegation; they are not always-on background services.
| Map and plan | Build and review | Verify and ship |
|---|---|---|
<img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f5fa.svg" alt="" aria-hidden="true" width="18"> Code Mapper (code_mapper) | <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f3a8.svg" alt="" aria-hidden="true" width="18"> Design Reviewer (design_reviewer) | <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/2705.svg" alt="" aria-hidden="true" width="18"> QA Lead (qa_lead) |
<img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f4da.svg" alt="" aria-hidden="true" width="18"> Docs Researcher (docs_researcher) | <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f9f0.svg" alt="" aria-hidden="true" width="18"> DevEx Auditor (devex_auditor) | <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/26a1.svg" alt="" aria-hidden="true" width="18"> Performance Auditor (performance_auditor) |
<img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f9ed.svg" alt="" aria-hidden="true" width="18"> Context Architect (context_architect) | <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f575.svg" alt="" aria-hidden="true" width="18"> Root-Cause Debugger (root_cause_debugger) | <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f50e.svg" alt="" aria-hidden="true" width="18"> Google SEO Auditor (google_seo_auditor) |
<img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/270d.svg" alt="" aria-hidden="true" width="18"> Prompt Architect (prompt_architect) | <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f4dd.svg" alt="" aria-hidden="true" width="18"> Docs Author (docs_author) | <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f6e1.svg" alt="" aria-hidden="true" width="18"> Security Auditor (security_auditor) |
<img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f50c.svg" alt="" aria-hidden="true" width="18"> MCP Integrator (mcp_integrator) | <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f4cb.svg" alt="" aria-hidden="true" width="18"> Spec Author (spec_author) | <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f9ea.svg" alt="" aria-hidden="true" width="18"> Test Verifier (test_verifier) |
<img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f3af.svg" alt="" aria-hidden="true" width="18"> Product Strategist (product_strategist) | <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f50d.svg" alt="" aria-hidden="true" width="18"> Code Reviewer (code_reviewer) | <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f6a2.svg" alt="" aria-hidden="true" width="18"> Release Verifier (release_verifier) |
<img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f3d7.svg" alt="" aria-hidden="true" width="18"> Engineering Planner (engineering_planner) | <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f3ad.svg" alt="" aria-hidden="true" width="18"> Frontend Verifier (frontend_verifier) | <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1fa7a.svg" alt="" aria-hidden="true" width="18"> Codex Doctor (codex_doctor) |
| Set | Skills |
|---|---|
| <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f373.svg" alt="" aria-hidden="true" width="20"> Local plugin | codex-chef-operator, offline-diagram-triplet, context-budget-planner |
| <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f9f0.svg" alt="" aria-hidden="true" width="20"> Reviewed global catalog | dependency-upgrade, gh-fix-ci, systematic-debugging, request-refactor-plan, security-best-practices, frontend-skill, webapp-testing, web-quality-audit, seo, accessibility, test-driven-development, documentation-and-adrs, mcp-builder, ai-project-starter, prompt-architect, ai-skill-create |
| Status | MCPs | Boundary |
|---|---|---|
| <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/2705.svg" alt="" aria-hidden="true" width="20"> Enabled by default | <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f4da.svg" alt="" aria-hidden="true" width="18"> OpenAI Docs · <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f9ed.svg" alt="" aria-hidden="true" width="18"> Context7 · <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f9e0.svg" alt="" aria-hidden="true" width="18"> Sequential Thinking · <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f3ad.svg" alt="" aria-hidden="true" width="18"> Playwright · <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f9f0.svg" alt="" aria-hidden="true" width="18"> Chrome DevTools · <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f5fa.svg" alt="" aria-hidden="true" width="18"> Serena · <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f9e9.svg" alt="" aria-hidden="true" width="18"> Memory | Research, code navigation, browser evidence, and non-secret local context. |
| <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f512.svg" alt="" aria-hidden="true" width="20"> Disabled until opt-in | <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f4c1.svg" alt="" aria-hidden="true" width="18"> Filesystem · <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f419.svg" alt="" aria-hidden="true" width="18"> GitHub · <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f3a8.svg" alt="" aria-hidden="true" width="18"> Figma · <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f4cc.svg" alt="" aria-hidden="true" width="18"> Linear · <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f5d2.svg" alt="" aria-hidden="true" width="18"> Notion · <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f6a8.svg" alt="" aria-hidden="true" width="18"> Sentry · ▲ Vercel · <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f5c4.svg" alt="" aria-hidden="true" width="18"> Supabase | These can expose private files, accounts, deployments, telemetry, or databases. |
Run npm run codex:status after install to see MCP setup notes, effective controls, routing profiles, and installed-runtime drift without mutating global Codex state.
<p align="center"> <img src="assets/workflow-overview.svg" alt="Workflow overview showing install, route, research, implement, and verify steps" width="100%" /> </p>
| Signal | Evidence |
|---|---|
| <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f9ea.svg" alt="" aria-hidden="true" width="20"> Real validation | npm run check runs repo, docs, install-plan, installer smoke, agent drift, MCP drift, token-surface, skill-source, supply-chain, package-surface, release-readiness, and security checks. |
| <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f310.svg" alt="" aria-hidden="true" width="20"> Multilingual surface | Six root READMEs plus six-language deep docs are validated. |
| <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f50c.svg" alt="" aria-hidden="true" width="20"> Conservative connectors | Account, database, production, and broad filesystem connectors stay disabled until a task explicitly needs them. |
| <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f9fe.svg" alt="" aria-hidden="true" width="20"> Install plan | manifests/install-plan.json and schemas/install-plan.schema.json define the managed write surface. |
| <img src="https://cdn.jsdelivr.net/gh/twitter/[email protected]/assets/svg/1f9ee.svg" alt="" aria-hidden="true" width="20"> Context budget | npm run token:audit reports the largest context surfaces; token-safe.config.toml lowers verbosity and tool-output budgets without disabling skills, MCPs, memory, hooks, or automatic agent model/reasoning selection. |
| Agent-readable index | llms.txt gives agents a compact map of install targets, docs, safety boundaries, and comparison sources. |
.github/ CI workflow plus issue and PR templates
assets/ Public-safe README visuals
catalog/ Agent, skill, MCP, and routing metadata
README*.md Multilingual public entry points
docs/ Six-language setup and verification guides
manifests/ No-write install plan metadata
plugins/ Bundled local Codex plugin and skills
schemas/ Lightweight validation schemas
scripts/ Install, doctor, status, CLI, and validation scripts
templates/codex/ Files copied into ~/.codex
templates/git/ Optional global Git hygiene filesnpm run check
git diff --check
gitleaks detect --redact --no-banner --no-git --verboseOptional online skill-source check:
npm run verify:skills:onlineUse the README for orientation; use the docs for operator detail:
AGENTS.md mapping: Research notesThis repo is built to be public-ready after validation, but the installer is local-only. Commit, push, tag, release, package publishing, deployments, and GitHub settings changes must be explicit human decisions after local verification.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.