Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/robzilla1738/codex-workflows/claude-mdgit clone --depth 1 https://github.com/robzilla1738/Codex-WorkflowsWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00524 | $0.00524 |
| Opus 5 | $0.00262 | $0.00262 |
| Sonnet 5 | $0.00105 | $0.00105 |
| Haiku 4.5 | $0.00052 | $0.00052 |
Grade A, and why
Codex-Workflows CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
This is a copy
95% identical to Codex-Workflows AGENTS.md — 2 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Claude instructions
This repo builds a Codex plugin, not a one-off demo. Treat generated artifacts, plugin metadata, and install docs as part of the product.
Project shape
- Source packages live under
packages/. - Built-in workflow templates live under
workflows/. - The installed plugin bundle is generated under
plugins/codex-workflows/. - Codex skill instructions live under
skills/codex-workflows/SKILL.md. - The repo marketplace is
.agents/plugins/marketplace.json.
Do not edit plugins/codex-workflows/dist/* directly. Change source files,
then run pnpm validate:plugin.
Commands
Use these before handing work back:
pnpm typecheck
pnpm test
pnpm validate:plugin
For a fast workflow smoke:
node plugins/codex-workflows/dist/cli.js validate workflows/bug-sweep.workflow.js --cwd /tmp
node plugins/codex-workflows/dist/cli.js run workflows/bug-sweep.workflow.js --cwd /tmp --adapter simulate --no-watch
For MCP smoke testing, start the generated server from
plugins/codex-workflows and call workflow_validate.
Distribution rules
- Keep install docs pointed at
robzilla1738/Codex-Workflows. - Keep plugin names stable: marketplace
codex-workflows, plugincodex-workflows. - Keep the plugin bundle self-contained. Users should not need to run
pnpmafter installing from GitHub. - Do not leave
/Users/...paths, local cache paths, or machine-specific binaries in the plugin bundle. - If package versions change, regenerate the plugin and commit the generated bundle.
Runtime rules
- Workflow scripts are declarative. They define phases and worker prompts.
- Workflow scripts must not get direct filesystem, shell, network, process, or Node built-in access.
- Bug-finding workflows should default to read-only workers.
- Model overrides from the caller beat workflow defaults.
- Default durable run state belongs under Codex home so read-only runs do not
dirty target repos. Project-local
.codex-workflows/runs/<run-id>/is opt-in and legacy-readable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 65 lines · 524 tokens per session scan A beee89ab2871
Codex-Workflows CLAUDE.md is an instructions file published in the GitHub repository robzilla1738/Codex-Workflows (63 stars, last pushed 2mo ago), licensed MIT. It adds 524 tokens to every session, about $0.0026 per session on Opus 5. A static security scan graded it A with 0 findings. It is 95% identical to Codex-Workflows AGENTS.md, differing in 2 lines, and is treated as a copy.
Other instructions, from other repositories
swarm-pi-code-plugin CLAUDE.md
Instructions for JiaWeiXie/swarm-pi-code-plugin: This repository treats documentation as part of the product change. Preserve unrelated dirty worktree changes and do not reset, stash, discard, commit, push, or install external tooling without explicit scope.
swarm-pi-code-plugin AGENTS.md
Instructions for JiaWeiXie/swarm-pi-code-plugin, a project described as: connects Claude Code and Codex to a bounded Pi coding worker for repository-grounded analysis, planning, review, and implementation. It is designed for teams that want a second coding-agent perspective without giving that worker unrestricted…
research-os AGENTS.md
Instructions for lxinfei5/research-os, covering researchos — constitution, §0 what this is, §1 directory discipline, §2 half-life (innovation 1) — memory design and §3 behavior pillars (innovations 2–5).
arc AGENTS.md
Instructions for howells/arc, covering arc, shape, gotchas, editing constraints and search order.
AI-System-Design-Consultant CLAUDE.md
Instructions for deepanshu2711/AI-System-Design-Consultant, covering claude.md, what this is, running it, load-bearing typos — do not "fix" these paths and agent node pattern.
agent-eve CLAUDE.md
Instructions for shehryarsaroya/agent-eve, covering claude.md — agent eve (né the compact, then agent transfer), 0. 🚨 hard rules, 1. what this is, 2. doc map and 3. state of play (2026-07-26).