codex-delegate · sateezg — Claude Code skill for delegating to Codex
The codex-delegate skill routes repetitive, high-volume coding tasks to Codex and guides Claude through checking the result. Use it for bulk edits, test generation, or repo-wide audits.
MIT ★ 195 updated 2026-08-10
/plugin marketplace add Sateezg/codex-bridge
/plugin install codex-bridge@codex-bridgeWhat codex-delegate does
codex-delegate helps Claude hand Codex work such as applying the same change across many files, creating fixtures, or inspecting routes for missing authorization checks. The Codex CLI uses the ChatGPT plan quota, keeping large intermediate results out of Claude’s context.
The skill has Claude define the scope and write a self-contained task brief before Codex starts. Claude then reviews the diff or runs tests and tells the user what it checked. It names codex-implementer for file edits, codex-reviewer for reviews, and codex-debugger for investigating failures; codex-run handles a quick question.
Codex has no access to the conversation history, so its brief needs the relevant files, rules, and acceptance criteria. Small edits and work that depends on the discussion stay with Claude.
What it helps with
- delegate repetitive edits across many files
- generate fixtures or a set of similar tests
- locate every call to a deprecated API
- review a diff produced by Codex
- write a self-contained task brief for a Codex subagent
Example prompts for the agent
- Hand the bulk symbol rename to Codex, then inspect the diff.
- Ask Codex to write tests for this module and check what it returns.
- Delegate the route authorization audit to Codex, then verify its findings.
How to install codex-delegate
1Plugin marketplace
/plugin marketplace add Sateezg/codex-bridge/plugin install codex-bridge@codex-bridgeThis installs the whole codex-bridge plugin with all its skills
2Manually: put the files in place
~/.claude/skills/codex-delegate/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/codex-delegate/
Manually: put the files in place
.agents/skills/codex-delegate/SKILL.mdFor all projects use ~/.agents/skills/
Manually: put the files in place
.cursor/skills/codex-delegate/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install Sateezg/codex-bridge skills/codex-delegate2Manually: put the files in place
.github/skills/codex-delegate/SKILL.mdPlatform CLI
gemini skills install https://github.com/Sateezg/codex-bridge --path skills/codex-delegateManually: put the files in place
.devin/skills/codex-delegate/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about codex-delegate
How do I install codex-delegate in Claude Code?
In Claude Code, run /plugin marketplace add Sateezg/codex-bridge, then /plugin install codex-bridge@codex-bridge.
What does codex-delegate do?
It helps Claude select suitable work for Codex, prepare a task brief, and review the result. It also guides Claude to keep small or context-dependent tasks.
What should a Codex task brief include?
State the goal, exact files or file pattern, the change rule, what must remain untouched, and the expected output. Codex cannot see the conversation history.
Source text
SKILL.md · MIT
--- description: Offer to hand heavy work to Codex (GPT-5) as a subagent so it runs on the ChatGPT plan quota instead of burning Claude context and tokens. Use at the START of any task that is large, repetitive, or asset-producing — bulk refactors across many files, generating boilerplate or fixtures, exhaustive repo-wide audits, writing many similar tests, and anything needing mockups or images. Also use whenever the user says "save tokens", "use codex", "delegate this", or "who should do this". --- # Delegate to Codex when it saves the user tokens The Codex CLI on this machine runs on the user's **ChatGPT plan quota** — a separate budget from this Claude session. Work pushed to Codex costs no Claude output tokens and, more importantly, keeps large intermediate output (file dumps, generated boilerplate, long search results) out of this context window. Your job is to notice when that trade is worth making, **offer it**, and then orchestrate. ## The decision rubric Delegate to Codex when the task is **high-volume and low-judgment**: | Signal | Example | | :-- | :-- | | Mechanical edits across many files | rename a symbol in 40 files, migrate an import style, add a header to every module | | Bulk generation | scaffold 20 CRUD endpoints, write fixtures, generate a large seed dataset | | Exhaustive sweeps | "find every place we call the old API", audit all routes for auth | | Assets | any image, icon set, mockup, or diagram — always Codex, it has gpt-image-2 | | Long, self-contained subtasks | port a module to TypeScript, write the test suite for one file | | A genuinely independent opinion | you already have an answer and want it challenged | Keep it yourself when the task is **low-volume and high-judgment**: architecture calls, ambiguous requirements, anything needing the conversation history, small edits (under ~3 files), or anything the user is iterating on interactively. Round-tripping a two-line fix through Codex is slower and costs *more* overall. ## How to offer it When the rubric says delegate, say so in one or two sentences **before** starting, and be concrete about the split. Don't ask a vague "should I use Codex?" — the user can't judge that. Tell them what each side does and what it saves: > This touches ~35 files with the same mechanical change. I can hand the edits to > Codex — it runs on your ChatGPT quota and keeps 35 file dumps out of this > context — then review its diff here. Want me to? If the user has already said "use Codex" or "save tokens", skip the question and just do it. If the task is small, don't offer at all — the offer itself is noise. ## How to orchestrate You stay the orchestrator. The pattern is always: **you scope, Codex executes, you verify.** 1. **Scope it yourself.** Work out exactly what needs doing — the file list, the rule, the acceptance criteria. This is the judgment part and it's cheap. 2. **Write a self-contained brief.** Codex has none of this conversation. Include the goal, the exact files or glob, the rule to apply, what must not change, and the output you expect back. 3. **Pick the right delegate:** - `codex-implementer` subagent — for edits to files (runs `workspace-write`) - `codex-reviewer` subagent — for review of a diff or module - `codex-debugger` subagent — for root-causing a failure - `codex-artist` subagent — for images, icons, mockups - `codex-run "<task>"` in Bash — for a single quick question, no subagent needed 4. **Verify what comes back.** Read the diff, run the tests, spot-check the images. Never forward Codex's claim of success without checking. This is the part that makes delegation safe, and it's why you don't delegate judgment. 5. **Report the split.** Tell the user what Codex did and what you checked. ## Guardrails - **Never delegate a task you can't verify.** If you couldn't tell a good result from a bad one, do it yourself. - **`workspace-write` only when file changes are the point**, and tell the user before Codex writes to their repo. Default everything else to read-only. - **Uncommitted work:** ask the user to commit or stash first when Codex will write to files, so its changes are reviewable as a clean diff. - **Don't fan out.** One Codex task at a time unless the user asked for parallel work; each one spends their quota. - **Codex is not always right.** It has no conversation context and can misunderstand a brief. Treat its output as a draft from a capable stranger.
© Sateezg · MIT · Sateezg/codex-bridge
Similar skills
All skills for Testing and QA →codex-delegate by other authors comes first.