What is kiri CLAUDE.md?

Instructions for working in the CAPHTECH/kiri project, including required use of KIRI's code-search and code-reading tools.

They help agents understand the codebase, search for files and patterns, read selected code sections, and trace dependencies.

They make the project's preferred exploration workflow clear and help agents find relevant code without scanning everything.

How CAPHTECH/kiri configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to CAPHTECH/kiri. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/CAPHTECH/kiri/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/CAPHTECH/kiri

Made for: Claude Code.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/caphtech/kiri/claude-md/github.svg)](https://agentmods.dev/instructions/caphtech/kiri/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/caphtech/kiri/claude-md"><img src="https://agentmods.dev/badge/instructions/caphtech/kiri/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/caphtech/kiri/claude-md"><img src="https://agentmods.dev/badge/instructions/caphtech/kiri/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 4,012 This file is loaded in full into every session.

When invoked 4,012 The same file — it is already loaded in full.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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 5.1	$0.04012	$0.04012
Opus 5.5	$0.01605	$0.01605
Sonnet 5.5	$0.00802	$0.00802
Haiku 4.5	$0.00401	$0.00401

Measured 2mo ago against content hash c26962095aa3, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

kiri 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 2mo 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.

CLAUDE.md · 401 lines

How it starts

The opening of the file, as written. The whole thing — 401 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Use KIRI MCP Tools - MANDATORY

This project uses KIRI MCP. You MUST proactively use KIRI MCP tools for all code exploration and search tasks.

When to Use KIRI Tools

ALWAYS use KIRI tools instead of Glob/Grep for:

Understanding codebase structure → Use mcp__kiri__context_bundle
- Finding relevant code for implementing features
- Understanding how existing features work
- Locating code related to specific functionality
- Example: mcp__kiri__context_bundle with goal "How does the daemon lifecycle work?"
Searching for files or code patterns → Use mcp__kiri__files_search
- Finding files by keywords or functionality
- Locating implementation patterns
- Example: mcp__kiri__files_search with query "daemon startup"
Reading specific code sections → Use mcp__kiri__snippets_get
- Reading only relevant parts of large files
- Minimizing token usage
- Example: mcp__kiri__snippets_get for path "src/daemon/daemon.ts"
Understanding dependencies → Use mcp__kiri__deps_closure
- Finding what files depend on a module (inbound)
- Finding what a module depends on (outbound)
- Impact analysis for refactoring
- Example: mcp__kiri__deps_closure with path "src/server/runtime.ts"
Refining search results → Use mcp__kiri__semantic_rerank
- Prioritizing files by semantic relevance
- Narrowing down search results

Performance Benefits

Token efficiency: KIRI returns only relevant snippets, not entire files
Precision: Context-aware search beats simple pattern matching
Speed: Indexed search is faster than filesystem traversal
Structure-aware: Understands code symbols and dependencies

Token-Saving Workflow (v0.9.6+)

Run files_search or context_bundle with compact: true to gather lightweight candidates.
Use snippets_get with compact: true when you only need metadata, or includeLineNumbers: true when copying exact line references (format: 1375→export function ...).
Request includeTokensEstimate: true on context_bundle only when you explicitly need the projected token count; it is skipped by default.

Read the full file on GitHub · 401 lines

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.

2mo ago First seen · 401 lines · 4,012 tokens per session scan A c26962095aa3

Subscribe to this mod's changes

kiri CLAUDE.md is an instructions file published in the GitHub repository CAPHTECH/kiri (153 stars, last pushed 8mo ago), licensed MIT. It adds 4,012 tokens to every session, about $0.0160 per session on Opus 5.5. 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.

kiri CLAUDE.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
kiri CLAUDE.mdCAPHTECH/kiri	Instructions file	153	8mo ago	MIT	A	4,012	Claude Code
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	9,719	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	today	MIT	A	7,461	Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	5,516	GitHub Copilot
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	today	MIT	A	4,680	Codex, OpenCode
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	3,093	GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness	Instructions file	275k	3d ago	MIT	A	4,313	Codex, OpenCode

kiri: Instructions file for Claude Code