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/markc/aitxt/claude-mdgit clone --depth 1 https://github.com/markc/aitxtWhat 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 | $0.00449 | $0.00449 |
| Opus 5 | $0.00225 | $0.00225 |
| Sonnet 5 | $0.00090 | $0.00090 |
| Haiku 4.5 | $0.00045 | $0.00045 |
Grade A, and why
aitxt 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.
What it actually says
CLAUDE.md
Guidance for Claude Code when working in this repository.
Repository
This repository hosts the ai.txt specification at https://aitxt.ing.
/
├── index.html # Documentation site
├── ai.txt # Canonical specification
├── llms.txt # Pointer to the spec for llms.txt-aware tools
├── robots.txt # Content-Signal + sitemap
├── sitemap.xml
├── README.md # GitHub entry point
├── CNAME # Domain config
├── LICENSE # CC0 public domain
└── .claude/skills/aitxt/
├── SKILL.md # Claude Code /aitxt skill
└── aitxt # Discovery script (bash), bundled with the skill
Claude Code Tooling
The /aitxt skill discovers and displays ai.txt files. It is a project skill,
so it is available automatically when Claude Code runs inside this repo. To
have it everywhere, copy the skill directory into your user skills:
cp -r .claude/skills/aitxt ~/.claude/skills/
The skill calls its bundled script via ${CLAUDE_SKILL_DIR}/aitxt; nothing
else needs installing. Restart Claude Code to load it.
Usage
/aitxt # Current directory
/aitxt ~/Dev/myproject # Local path
/aitxt https://example.com # Remote URL
The script cascades up the directory/URL tree to find ai.txt, follows linked ai.txt files (depth=1), and displays a graph of discovered contexts.
Editing Guidelines
- The canonical spec is
/ai.txt— keep it minimal and self-contained - No markdown links to external ai.txt files in the spec (tools auto-follow any
[..](..ai.txt)link); bare backticked paths are fine - Documentation belongs in
index.html, not the spec - All frontmatter fields remain optional
- Bump
updated:inai.txtandlastmodinsitemap.xmlwhen the spec changes
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 · 54 lines · 449 tokens per session scan A e16532b39629
aitxt CLAUDE.md is an instructions file published in the GitHub repository markc/aitxt (5 stars, last pushed 8d ago), licensed CC0-1.0. It adds 449 tokens to every session, about $0.0022 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
modelcontextprotocol AGENTS.md
AGENTS.md instructions for modelcontextprotocol/modelcontextprotocol, covering model context protocol (mcp), contribution policy for ai agents, documentation structure, documentation guidelines and specification versioning.
modelcontextprotocol CLAUDE.md
Claude Code instructions for modelcontextprotocol/modelcontextprotocol, a project described as: Specification and documentation for the Model Context Protocol.
adl AGENTS.md
Instructions for inference-gateway/adl, covering agents.md, commands, schema versioning contract — the most important rule, style and testing.
agentnative AGENTS.md
AGENTS.md instructions for brettdavies/agentnative, covering agent instructions, project, authoritative content, voice and editing principles.
adl CLAUDE.md
Instructions for inference-gateway/adl, a project described as: A declarative language for defining AI agents, their capabilities, skills and tools. ADL simplifies agent development by generating consistent, enterprise-ready code from a single manifest. Build faster. Deploy smarter. Stay consistent.
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.