agentleFS
Sign inStart free
Ansenchen123Ansenchen123/claude-codex-workflow/skills/codex-dispatch/SKILL.md

codex-dispatch

Skill1 starsChanged 4 months agoMITView on GitHub

Use this skill whenever implementation work should be dispatched to Codex CLI instead of being written directly by Claude.

5 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions5 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    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 repositoryOne other file it gives its agents.

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 4 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 5 edited versions we read keep these.

  • 4

    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
  • could be automatic: a setting could enforce it
  1. Distinctive:

    Never write implementation code yourself.

    line 8
  2. Distinctive:

    Keep each task independently implementable and give it one verification command.

    line 27
  3. Distinctive:

    Run the command from the target repository so the scripts automatically target the current repo:

    line 31
  4. Distinctive:

    If the task passes, stage only the reviewed files for that task, inspect the staged diff, and create one commit.

    line 42

“Common” is measured: kept by at least 6 in 10 of the 5 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Codex Dispatch

Never write implementation code yourself. Plan the work, dispatch it to Codex CLI, and review the result only.

Plan

From the target repository, resolve its git toplevel and write each task spec there as tasks/TNNN-slug.md. Use this exact template copied from WORKFLOW.md:

# TNNN-slug
## Goal
(1–3 sentences)
## Acceptance criteria
(bullet list, each objectively checkable by command)
## Constraints
- Work strictly inside this repo; no global installs; do not modify tasks/, reports/, WORKFLOW.md, dispatch.ps1, dispatch.sh
(task-specific constraints)
## Verification command
(one re-runnable command; Codex must actually run it, Claude re-runs it during review)

Keep each task independently implementable and give it one verification command.

Dispatch

Run the command from the target repository so the scripts automatically target the current repo:

  • Windows: powershell.exe -NoProfile -ExecutionPolicy Bypass -File "${CLAUDE_PLUGIN_ROOT}/dispatch.ps1" TNNN-slug
  • macOS/Linux: bash "${CLAUDE_PLUGIN_ROOT}/dispatch.sh" TNNN-slug

The scripts run Codex sandboxed with --sandbox workspace-write by default. Use -Unsafe on Windows or --unsafe on macOS/Linux only when deliberately opting out.

Review

Review only reports/TNNN-slug.md and the git diff, including git status --short for untracked files. Re-run the verification command from the spec; never rely only on Codex's report.

If the task passes, stage only the reviewed files for that task, inspect the staged diff, and create one commit; never use git add -A in a dirty worktree. If it fails, append ## Revision feedback (Rn) to the same spec with the factual gaps, then re-dispatch it.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.

Check your setup, free