Sateezg/codex-bridge
Image generation (gpt-image-2) and GPT-5 subagents for Claude Code — through the Codex CLI login you already have. No OpenAI API key.
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
Copy this rig
# review before running: this installs third-party code $ npx degit Sateezg/codex-bridge/agents ./rig-codex-bridge/agents $ npx degit Sateezg/codex-bridge/skills ./rig-codex-bridge/skills
MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.
$ npx degit Sateezg/codex-bridge/skills/ask-codex .claude/skills/ask-codex $ npx degit Sateezg/codex-bridge/skills/asset-set .claude/skills/asset-set $ npx degit Sateezg/codex-bridge/skills/codex-delegate .claude/skills/codex-delegate $ npx degit Sateezg/codex-bridge/skills/codex-review .claude/skills/codex-review $ npx degit Sateezg/codex-bridge/skills/edit-image .claude/skills/edit-image $ npx degit Sateezg/codex-bridge/skills/generate-image .claude/skills/generate-image
$ curl -fsSL --create-dirs -o .claude/agents/codex-artist.md https://raw.githubusercontent.com/Sateezg/codex-bridge/main/agents/codex-artist.md $ curl -fsSL --create-dirs -o .claude/agents/codex-debugger.md https://raw.githubusercontent.com/Sateezg/codex-bridge/main/agents/codex-debugger.md $ curl -fsSL --create-dirs -o .claude/agents/codex-implementer.md https://raw.githubusercontent.com/Sateezg/codex-bridge/main/agents/codex-implementer.md $ curl -fsSL --create-dirs -o .claude/agents/codex-reviewer.md https://raw.githubusercontent.com/Sateezg/codex-bridge/main/agents/codex-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/codex-second-opinion.md https://raw.githubusercontent.com/Sateezg/codex-bridge/main/agents/codex-second-opinion.md
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 7,204 rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(rm -rf *)",
"Read(**/*.pem)",
"Bash(rm -rf /)",
"Bash(git push --force:*)",
"Bash(sudo *)",
"Read(.env)",
"Bash(rm -rf /*)",
"Read(./.env.*)",
"Read(~/.aws/**)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Read(**/*.key)",
"Read(**/.env.*)",
"Bash(sudo:*)",
"Bash(git reset --hard*)",
"Bash(git reset --hard:*)",
"Read(.env.*)"
],
"ask": [
"Bash(git push:*)",
"Bash(git push *)",
"Bash(git commit:*)",
"Bash(rm *)",
"Bash(rm:*)",
"Bash(git rebase *)",
"Bash(wget *)",
"Bash(npm publish:*)",
"Bash(git commit *)",
"Bash(gh pr merge *)"
]
}
} Skills (6)
Subagents (5)
| codex-artist | Image-generation subagent. Use PROACTIVELY whenever a task needs image files — generating or editing icons, logos, banners, illustrations, mockups, photos, or a full asset set (favicons, OG cards, app |
| codex-debugger | Root-cause a failing test, crash, stack trace, or misbehaving feature by delegating the investigation to OpenAI Codex (GPT-5) via the local Codex CLI, then verifying its diagnosis against the real cod |
| codex-implementer | Hand a well-specified, high-volume coding task to OpenAI Codex (GPT-5) to actually write the code, so it runs on the user's ChatGPT quota instead of this session's context. Use for mechanical work acr |
| codex-reviewer | Independent code reviewer powered by OpenAI Codex (GPT-5) via the local Codex CLI. Use when the user asks to review changes, a PR, a branch, or a module — especially before a commit or merge, or when |
| codex-second-opinion | Get an independent opinion from OpenAI Codex (GPT-5) on a design decision, an approach, a tradeoff, or an unfamiliar area of the codebase — a general-purpose Codex liaison for questions that don't fit |
Similar rigs
jessepike/claude-codex-handoff
multi-model handoff pattern for Claude Code + Codex
Minimalist 990 tok ·
jnurre64/claude-skill-image-gen
Claude Code skills for generating images via ComfyUI, plus a Windows setup skill. Independent, community-built; not affiliated with Anthropic or Comfy Org.
Minimalist 178 tok ·
chaseai-yt/claudex-loop
Claude Code skill: four-phase plan hardening (recon, interrogate, Codex adversarial review, cross-model build & inspection) — two AI models harden your plan before a line of code exists, then swap jobs to build it. Whoever built it never gr
Minimalist 272 tok ·
MysticalNobody/alatyr_flutter
A hardened Flutter starter and GitHub template for AI-agent development, with machine-enforced architecture, automated quality gates, and a Claude Code + Codex implementation/review workflow
Pragmatist 2.4k tok ·