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/yanchuk/claude-plugin-codex/agents-mdgit clone --depth 1 https://github.com/yanchuk/claude-plugin-codexWhat 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 | $0.00537 | $0.00537 |
| Opus 5 | $0.00269 | $0.00269 |
| Sonnet 5 | $0.00107 | $0.00107 |
| Haiku 4.5 | $0.00054 | $0.00054 |
Grade A, and why
claude-plugin-codex AGENTS.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.
How it starts
The opening of the file, as written. The whole thing — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Instructions
This repo builds claude-plugin-codex, a Codex plugin that lets Codex consult
local Claude Code for advice, reviews, adversarial checks, and rescue tasks.
Codex remains the orchestrator.
Scope
- Work only in this standalone repo.
- Do not modify
/Users/yanchuk/Documents/GitHub/skill-arsenalfor this project. - Do not commit or push unless the latest user instruction explicitly asks for it.
- Keep plugin metadata public-release ready:
.codex-plugin/plugin.json,.agents/plugins/marketplace.json, andREADME.mdshould stay consistent. - Keep
package.jsonandplugins/claude-code-advisor/.codex-plugin/plugin.jsonon the same version.
Commands
- Run
npm run validatebefore claiming the plugin is ready. - Run
npm run test:smokeafter changes that touch Claude CLI invocation, runtime behavior, or install instructions. - Run
npm run test:e2e:codexafter changes that affect Codex plugin routing, skill instructions, or public install behavior.
Claude Model Policy
- Do not default Claude advisor, review, adversarial-review, rescue, or monitor work to Sonnet.
- Let Claude Code use the user's configured default model unless the user explicitly asks for another model.
- Pass xhigh effort for Claude advisor, review, adversarial-review, rescue, and background work unless the user explicitly asks for another effort.
- Use Sonnet only for junior-agent delegation governed by the
tasks-for-sonnetskill, where Sonnet acts as Hands, scout, verifier, or synthesis worker.
Documentation
- Put public install instructions near the top of
README.md. - Write README and marketplace copy in plain, direct language.
- Do not document unsupported Codex plugin manifest fields as real features.
$claudeis the stable skill command./claudeand/claude:rescueare best-effort aliases only when Codex passes slash-style text through to the skill.
Context7
Use the ctx7 CLI for current documentation when working from docs for a
library, framework, SDK, API, CLI tool, or cloud service:
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 · 60 lines · 537 tokens per session scan A c9ce23721413
claude-plugin-codex AGENTS.md is an instructions file published in the GitHub repository yanchuk/claude-plugin-codex (18 stars, last pushed 3mo ago), licensed MIT. It adds 537 tokens to every session, about $0.0027 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
building-a-coding-agent-from-scratch-course AGENTS.md
Instructions for decodingai-magazine/building-a-coding-agent-from-scratch-course, covering decode, project structure, tech stack, docs & external services and running commands.
dev-agent-skills CLAUDE.md
Instructions for fvadicamo/dev-agent-skills, covering claude.md, project overview, architecture, key file: marketplace.json and skill anatomy.
toki AGENTS.md
Instructions for choi138/toki, covering workspace rules, source of truth, toki workflow, required checks and global safety rules.
timps AGENTS.md
Instructions for Sandeeprdy1729/timps, covering agents.md — timps codebase guide, repo layout, typescript config drift, agent entrypoints (timps-code) and auto-capture & context-anywhere (timps setup / recall).
kmux AGENTS.md
Instructions for kkd927/kmux, covering agents.md, kmux focus and test discipline.
surreal-skills AGENTS.md
Instructions for 24601/surreal-skills, covering agent briefing: surrealdb, quick start, check if configured, get full capabilities manifest (json) and capabilities.