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/shenchangmin/cairnkit/claude-mdgit clone --depth 1 https://github.com/shenchangmin/cairnkitWrote 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/shenchangmin/cairnkit/claude-md)<a href="https://agentmods.dev/instructions/shenchangmin/cairnkit/claude-md"><img src="https://agentmods.dev/badge/instructions/shenchangmin/cairnkit/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 | $0.01595 | $0.01595 |
| Opus 5 | $0.00797 | $0.00797 |
| Sonnet 5 | $0.00319 | $0.00319 |
| Haiku 4.5 | $0.00160 | $0.00160 |
Grade A, and why
cairnkit 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 — 149 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Guidance for AI coding agents and humans working in the cairnkit repository. This is the repository operating contract for the Claude Code harness; AGENTS.md is the equivalent Codex-facing entry. Both describe the same engine and shared role content.
cairnkit is a knowledge-precipitation delivery harness for Claude Code and Codex. The workflow is the pipe; knowledge is the moat.
First principle
The durable product is the knowledge loop: inject reviewed knowledge before work, record its use, extract transferable learning after work, and move it across projects only through explicit review and publication boundaries. The staged delivery workflow exists to make that loop reliable.
When workflow convenience conflicts with knowledge quality or admission safety, knowledge quality wins. Do not turn cairnkit into a prompt-only workflow whose state or gates live in model context.
Architecture and sources of truth
Claude Code commands/agents ─┐
├─ shared commands/ + agents/ + skills/
Codex native skills plugin ──┘ │
▼
cairn Rust engine (deterministic)
│
▼
STATE.yaml + workflow/knowledge files
cairnis the only writer of.cairnkit/STATE.yaml; never edit it by hand.- Stage gates, path validation, lifecycle rules, shared-KB admission, and Git publication checks belong in Rust. Analysis, design, and role mandates belong in Markdown.
- Files are the inter-role protocol. Each role worker reads files and the engine state, writes only its declared artifact, and never relies on another role's conversation history.
- Claude Code and Codex are adapters, not separate products. Changes to shared roles or commands must remain behaviorally identical in both forms.
/cairnkit:evolverecords proposals only. Applying a proposal still requires human review and never implies a commit, push, notification, or deployment.
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 · 149 lines · 1,595 tokens per session scan A 73721fd5b5a6
cairnkit CLAUDE.md is an instructions file published in the GitHub repository shenchangmin/cairnkit (1 stars, last pushed 1mo ago), licensed MIT. It adds 1,595 tokens to every session, about $0.0080 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
llm-wiki-plugin AGENTS.md
AGENTS.md instructions for praneybehl/llm-wiki-plugin, covering repository instructions and releases.
kaas CLAUDE.md
Instructions for bybit-exchange/kaas, covering claude.md, behavioral guidelines, 1. think before coding, 2. simplicity first and 3. surgical changes.
Zettelgarden CLAUDE.md
Instructions for Zettelgarden/Zettelgarden, covering claude.md, project overview, development commands, frontend (zettelkasten-front/) and backend (go-backend/).
Zettelgarden AGENTS.md
Instructions for Zettelgarden/Zettelgarden, covering repository guidelines, project structure & module organization, build, test, and development commands, coding style & naming conventions and testing guidelines.
data-olympus AGENTS.md
Instructions for knaisoma/data-olympus, covering data olympus, project overview, architecture, development workflow and code standards.
dev-loop AGENTS.md
AGENTS.md instructions for choiyounggi/dev-loop, covering dev-llm-wiki — agent schema, what this wiki is, directory layout, routing protocol (how to consume) and page format (how to write).