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/runnithan/claudex-setup/claude-mdgit clone --depth 1 https://github.com/runnithan/claudex-setupWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/runnithan/claudex-setup/claude-md)<a href="https://agentmods.dev/instructions/runnithan/claudex-setup/claude-md"><img src="https://agentmods.dev/badge/instructions/runnithan/claudex-setup/claude-md.svg" alt="Measured on agentmods" height="20"></a>What 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 | $0.02304 | $0.02304 |
| Opus 5 | $0.01152 | $0.01152 |
| Sonnet 5 | $0.00461 | $0.00461 |
| Haiku 4.5 | $0.00230 | $0.00230 |
Grade A, and why
claudex-setup 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 4d 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 — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.
claudex-setup
Portable Claude Code configuration for bootstrapping project-local agent instructions, commands, hooks, plugins, skills, and reference material.
Architecture
- Library docs: Context7 MCP is available for fetching up-to-date package documentation when training data may be outdated. GSD agents use it; available for general use too.
- Tool routing: design, planning, hook and subagent work goes to Claude Code (more hook events, and it auto-delegates subagents). Long plan execution, second-opinion review and research-heavy pulls go to Codex (far fewer output tokens for the same work). Route by task type rather than picking a winner.
AGENTS.mddefers to this file rather than restating it, so this line is the single source.
Project structure
See references/project-structure.md for the full directory tree.
- Transcript → lessons pipeline: YouTube transcripts land in
transcripts/(fetched by theYouTube URL Updaterscheduled job, scripts inscripts/). Run/extract-lessons(.claude/commands/extract-lessons.md) to mine them into one-file-per-lesson underlessons/<category>/, indexed in lessons/INDEX.md and dedup-tracked bylessons/.processed.json. Don't hand-place lessons or copy lesson content into CLAUDE.md. Perlessons/README.md, CLAUDE.md stays curated (promote a proven rule manually).
Workflow
- PRs target
main. - Commit messages follow Conventional Commits:
<type>(<optional scope>): <subject>, imperative mood, lowercase subject, no trailing period. Allowed types:feat,fix,docs,style,refactor,perf,test,build,ci,chore,revert. Use the body to explain why; add aBREAKING CHANGE:footer for incompatible changes. Examples:feat(transcripts): add daily fetch job,docs(projects): add entries index,fix(fetch): cap runs to avoid YouTube throttle. - Commits are authored solely by the repository owner, so omit Co-Authored-By lines. This preserves clean authorship in the public commit history.
- After merging a PR, the feature branch auto-deletes (repo setting).
- Keep
CHANGELOG.mdcurrent, don't let it drift. Any user-facing change (new/changed commands, agents, skills, hooks, pipeline behaviour, safety/permission changes, notable lessons-pipeline additions) gets a Keep a Changelog entry (Added/Changed/Fixed) in the same change, not batched up "later." On release, bump the version inCHANGELOG.md,.claude-plugin/plugin.json, and.claude-plugin/marketplace.jsontogether. - Write CHANGELOG entries generically: never name a private project in them.
CHANGELOG.mdis not inpublish.sh's scrub set (unlike.claude/commands/optimise.md/references/project-structure.md, where private project names are auto-genericized), so a private name written here is not scrubbed and the public leak-gate hard-fails the publish on it. Describe the feature, not the project that prompted it, e.g./optimise <project> --design, not the real project name.
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.
- 4d ago First seen · 83 lines · 2,304 tokens per session scan A d71532fe4bb3
claudex-setup CLAUDE.md is an instructions file published in the GitHub repository runnithan/claudex-setup (1 stars, last pushed 3d ago), licensed MIT. It adds 2,304 tokens to every session, about $0.0115 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 2026-08-31.
Other instructions, from other repositories
river-review GEMINI.md
Gemini CLI instructions for s977043/river-review, covering gemini instructions (river-review), 必須チェック, gemini-specific tips, prompt assembly and development tips.
living-manual CLAUDE.md
Instructions for dougstanford/living-manual, covering living-manual, start of every session, how work lands — non-negotiable, user's manual (living-manual) and releasing.
ship-it open-pr.instructions.md
Instructions for khrichtchatyi/ship-it: When the user asks to open, create, or submit a pull request, follow the skill in skills/open-pr/SKILL.md in full: verify gh is authenticated, resolve the target repository and default branch, push the current branch, open one pull request with a title and body derived from the…
agent-skills CLAUDE.md
Instructions for mindbox-cloud/agent-skills, covering agent-skills — developer guide for claude code, repository purpose, repository structure, how to add a new plugin and step 1 — create the directory structure.
my-claude-skills CLAUDE.md
Instructions for brody-0125/my-claude-skills, covering claude code skills repository, validation commands, sub-kopring-engineer: kotlin/java 코드 컨벤션 검증, numerical: 수치 연산 코드 검증 and sub-test-engineer: 문서 일관성 검증.
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).