codex-delegate
Skill248 starsChanged 3 months agoMITView on GitHub
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".
15 other projects use a version of this.
Adopted by
Most recently taken up by
- tzoofu/skillstheir own version, first seen Nov 8, 2026
- Song-Sungyoub/claude-codex-delegatetheir own version, first seen Nov 8, 2026
- antorsae/dual-agenttheir own version, first seen Nov 1, 2026
- lnopy96/claude-skillstheir own version, first seen Nov 1, 2026
- antonailog/skillstheir own version, first seen Nov 1, 2026
- takayoshitoyoda05/claude-ml-templatetheir own version, first seen Oct 20, 2026
Other versions15 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 15
- amElnagdy/delegate-skills★ 2.6ksame name
- kachofugetsu09/akashic-agent★ 520same name
- WenyuChiou/codex-delegate★ 62same name
- antorsae/dual-agent★ 17same name
- gigio1023/agent-skills★ 2same name
- gigio1023/gigio-skills★ 2same name
- Song-Sungyoub/claude-codex-delegate★ 2same name
- thesawdawg/agent-skills★ 2same name
- aashahin/claude-codex-delegate★ 1same name
- antonailog/skills★ 1same name
5 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository5 other files it gives its agents.
- ask-codexSkill
- asset-setSkill
- codex-reviewSkill
- edit-imageSkill
- generate-imageSkill
The file
We split this skill into the rules it gives an agent
7 rules. 0 are common, 7 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 11 edited versions we read keep these.
7
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 28
Keep it yourself when the task is low-volume and high-judgment: architecture calls, ambiguous requirements, anything needing the …
- Distinctive: line 35
When the rubric says delegate, say so in one or two sentences before starting, and be concrete about the split.
- Distinctive: line 43
If the user has already said "use Codex" or "save tokens", skip the question and just do it.
- Distinctive: line 62
Verify what comes back: Read the diff, run the tests, spot-check the images.
- Distinctive: line 63
Never forward Codex's claim of success without checking.
- Distinctive: line 69
Never delegate a task you can't verify: If you couldn't tell a good result from a bad one, do it yourself.
- Distinctive: line 73
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.
“Common” is measured: kept by at least 6 in 10 of the 11 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
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.
- 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.
- 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.
- Pick the right delegate:
codex-implementersubagent — for edits to files (runsworkspace-write)codex-reviewersubagent — for review of a diff or modulecodex-debuggersubagent — for root-causing a failurecodex-artistsubagent — for images, icons, mockupscodex-run "<task>"in Bash — for a single quick question, no subagent needed
- 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.
- 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-writeonly 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.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.