{"id":51251456,"url":"https://github.com/heurema/codex-partner","last_synced_at":"2026-06-29T07:02:11.076Z","repository":{"id":340366324,"uuid":"1165642269","full_name":"heurema/codex-partner","owner":"heurema","description":"Codex CLI as second AI partner — review, debate, implement alongside Claude Code","archived":false,"fork":false,"pushed_at":"2026-02-25T15:23:39.000Z","size":12,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-02-26T15:27:57.206Z","etag":null,"topics":["ai-agent","anthropic","claude","claude-code","codex","dual-ai","gpt"],"latest_commit_sha":null,"homepage":null,"language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/heurema.png","metadata":{"files":{"readme":"README.md","changelog":"CHANGELOG.md","contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-02-24T11:40:26.000Z","updated_at":"2026-02-25T15:29:41.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/heurema/codex-partner","commit_stats":null,"previous_names":["real-ai-engineering/codex-partner","heurema/codex-partner"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/heurema/codex-partner","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/heurema%2Fcodex-partner","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/heurema%2Fcodex-partner/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/heurema%2Fcodex-partner/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/heurema%2Fcodex-partner/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/heurema","download_url":"https://codeload.github.com/heurema/codex-partner/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/heurema%2Fcodex-partner/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":34916411,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-06-29T02:00:05.398Z","response_time":58,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ai-agent","anthropic","claude","claude-code","codex","dual-ai","gpt"],"created_at":"2026-06-29T07:02:07.685Z","updated_at":"2026-06-29T07:02:11.066Z","avatar_url":"https://github.com/heurema.png","language":null,"funding_links":[],"categories":[],"sub_categories":[],"readme":"# Codex Partner — Second AI in the Team\n\nThis repository provides workflows and a Claude Code skill for using OpenAI Codex CLI as an independent verification AI alongside your primary AI agent. The pattern treats Codex not as a replacement but as a peer — a second model with different training, different reasoning, and different blind spots — that you consult at key decision points: before committing, when evaluating competing approaches, or when you want implementation delegated to an isolated environment.\n\n## Why Multi-Model Verification\n\n- Different models have different reasoning strengths and blind spots. What one model overlooks, another often catches.\n- Independent review finds issues that single-model workflows miss — especially in security and logic, where model-specific reasoning patterns create systematic gaps.\n- Style disagreements between models are noise, not signal. Two models trained differently will have different formatting preferences. Filter those out; only bugs, security vulnerabilities, and logic errors matter.\n- Two perspectives structured into a clear format — with explicit divergence analysis — give you better information than asking one model to review its own work.\n\n## Team Model\n\n```\nHuman (domain knowledge, final decisions)\n  ↕\nPrimary AI (orchestrator, deep codebase understanding)\n  ↕\nCodex CLI (independent verification, second opinion)\n```\n\nThe human stays in control. The primary AI orchestrates and understands context. Codex CLI provides an independent signal. Nothing gets applied automatically — every finding goes through the human.\n\n## Requirements\n\n- OpenAI Codex CLI installed:\n  ```bash\n  npm install -g @openai/codex\n  ```\n- OpenAI API key configured (`codex auth` or `OPENAI_API_KEY` env var)\n- Git repository (for implement mode and review mode)\n- Optional: Claude Code (for plug-and-play skill integration)\n\n## Install\n\n**For Claude Code users (one-liner):**\n\n```bash\ngit clone https://github.com/Real-AI-Engineering/codex-partner.git ~/.claude/skills/codex-partner \u0026\u0026 ln -s ~/.claude/skills/codex-partner/skills/codex-partner ~/.claude/skills/codex-partner-skill\n```\n\n**Verify:** restart Claude Code, then type `/codex` — the skill should activate and ask which mode you want (review, ask, or implement).\n\n**Update:**\n\n```bash\ncd ~/.claude/skills/codex-partner \u0026\u0026 git pull\n```\n\n**Uninstall:**\n\n```bash\nrm ~/.claude/skills/codex-partner-skill \u0026\u0026 rm -rf ~/.claude/skills/codex-partner\n```\n\n**For standalone use:** no install needed — read the `docs/` directory for workflow documentation you can adapt to any AI assistant.\n\n## Three Modes\n\n| Mode | Purpose | Command |\n|------|---------|---------|\n| Review | Independent code review of current changes | `/codex review` |\n| Ask | Get second opinion for debate | `/codex ask \"question\"` |\n| Implement | Delegate implementation to isolated worktree | `/codex implement \"spec\"` |\n\nDetailed documentation:\n- [Review mode](docs/review-mode.md)\n- [Ask mode](docs/ask-mode.md)\n- [Implement mode](docs/implement-mode.md)\n\n## Rules\n\nSeven guardrails that make this pattern safe and useful:\n\n1. **Never auto-apply Codex findings.** Always present to user first.\n2. **Filter style disagreements.** Different models have different aesthetics — only flag bugs, logic errors, security, and performance.\n3. **Max 2 Codex calls per invocation.** Each call has ~8K+ token overhead. If more are needed, invoke `/codex` again as a fresh call.\n4. **Cost awareness.** Mention token cost if the task is trivial for Codex — not every question warrants a full model call.\n5. **Error handling.** Report Codex CLI failures clearly (auth expired, network error, timeout) and suggest next steps.\n6. **Worktree cleanup.** Always clean up after implement mode — on success, failure, or discard. Never leave orphaned worktrees.\n7. **No Codex for Codex.** Do not use Codex to review Codex output. One round only.\n\n## See Also\n\nOther [Real-AI-Engineering](https://github.com/Real-AI-Engineering) projects:\n\n- **[sigil](https://github.com/Real-AI-Engineering/sigil)** — risk-adaptive development pipeline with adversarial code review\n- **[herald](https://github.com/Real-AI-Engineering/herald)** — daily curated news digest plugin for Claude Code\n- **[teams-field-guide](https://github.com/Real-AI-Engineering/teams-field-guide)** — comprehensive guide to Claude Code multi-agent teams\n- **[proofpack](https://github.com/Real-AI-Engineering/proofpack)** — proof-carrying CI gate for AI agent changes\n\n## Sources\n\n- [OpenAI Codex CLI](https://github.com/openai/codex) — Official CLI repository\n- [Codex CLI Documentation](https://platform.openai.com/docs/guides/codex) — OpenAI platform docs\n- [Multi-model verification patterns](https://github.com/Real-AI-Engineering/teams-field-guide) — Agent Teams field guide with orchestration patterns\n\n## Contributing\n\nSee [CONTRIBUTING.md](CONTRIBUTING.md).\n\n## License\n\nMIT — see [LICENSE](LICENSE).\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fheurema%2Fcodex-partner","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fheurema%2Fcodex-partner","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fheurema%2Fcodex-partner/lists"}