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/nrslib/takt/claude-mdgit clone --depth 1 https://github.com/nrslib/taktWrote 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/nrslib/takt/claude-md)<a href="https://agentmods.dev/instructions/nrslib/takt/claude-md"><img src="https://agentmods.dev/badge/instructions/nrslib/takt/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 | $0.03512 | $0.03512 |
| Opus 5 | $0.01756 | $0.01756 |
| Sonnet 5 | $0.00702 | $0.00702 |
| Haiku 4.5 | $0.00351 | $0.00351 |
Grade A, and why
takt 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 — 146 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.
This file is not user-facing documentation. It exists for Claude Code to read while working in the repository — it does not substitute for the real docs under
docs/orREADME.md. Content being recorded here NEVER implies "we don't need to write this up for users." Anything users should see must still be added toREADME.md/docs/**separately.
Project Overview
TAKT (TAKT Agent Koordination Topology) is a multi-agent orchestration CLI. It runs AI agents through YAML-defined workflows with rule-based routing, parallel review, dynamic task decomposition, and worktree-isolated execution. This repository dogfoods TAKT for its own review and development.
Development Commands
| Command | Notes |
|---|---|
npm run build |
tsc (+ opencode-probe tsconfig) plus copies of src/shared/prompts/{en,ja}/**/*.md, src/shared/i18n/*.yaml, and src/core/runtime/presets/*.sh into dist/. Skipping any copy breaks runtime resolution. |
npm run watch |
TypeScript incremental build (no asset copy). |
npm run lint |
ESLint on src/. no-explicit-any is error; unused vars must be prefixed _. |
npm test |
Fast unit gate: type-contracts tsc, the test type-check (tsconfig.tests.json), then an adaptive number of unit shards launched concurrently (capped at 8 from availableParallelism() in scripts/run-npm-test.mjs, each shard --maxWorkers=1). Excludes integration tests; the output names the follow-up command. |
npm run test:it |
Light integration gate for real filesystem, bounded storage, and multi-component contracts. Run after implementation. |
npm run test:it:heavy |
Full heavy integration gate for real child processes, Git, complete engines, integration/regression/performance suites, and serial groups. Local execution uses one worker; PR CI shards across isolated runners. |
npm test -- src/__tests__/<file>.test.ts |
Route a single file to the correct unit, light-IT, heavy-parallel-IT, or heavy-serial-IT runner. For an added or changed IT, also run releaseVerificationWiring.test.ts by itself. Always target-run an added or changed heavy IT before handoff. |
npm test -- -t "<pattern>" |
Run unit tests whose name matches <pattern>. |
npm run test:opencode-probe |
Deterministic OpenCode probe smoke gate (11 cases, no API cost). Standalone; not part of routine gates or check:release. |
npm run test:e2e:mock |
Full mock-provider E2E suite (parallel shards). Single spec: TAKT_E2E_PROVIDER=mock npx vitest run --config vitest.config.e2e.mock.ts e2e/specs/<file>.e2e.ts — without that env var the specs run against the real provider (provider: claude in e2e/fixtures/config.e2e.yaml) and cost API credits. |
npm run test:e2e:provider:{claude,claude-sdk,codex,opencode,cursor} |
E2E against a real provider (slow, costs API credits). |
npm run check:release |
Full pre-release gate: build + lint + adaptive local unit + light IT + heavy IT + e2e. The main pull-request CI workflow runs the unit gate separately in 8 isolated shard jobs; local check:release does not run that CI matrix. |
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 · 146 lines · 3,512 tokens per session scan A 755ecdcc9911
takt CLAUDE.md is an instructions file published in the GitHub repository nrslib/takt (1,328 stars, last pushed 2d ago), licensed MIT. It adds 3,512 tokens to every session, about $0.0176 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
dos-kernel AGENTS.md
Instructions for anthony-chaudhary/dos-kernel, covering what dos is (the 30-second version), run it yourself first — the 60-second proof, when the user asks you about dos (the consumer moves), build, test, and check your work and the layout in one screen.
dos-kernel CLAUDE.md
Instructions for anthony-chaudhary/dos-kernel, covering dos — the dispatch operating system, the layering — keep these apart, the syscall abi, install & test and dos on dos — dogfood the kernel here.
PuPu CLAUDE.md
Claude Code instructions for haoxiang-xu/PuPu, covering gitnexus — code intelligence, always do, never do, resources and cli.
dos-kernel GEMINI.md
Instructions for anthony-chaudhary/dos-kernel, covering dos — the part that doesn't believe the agents, the rule, why this matters and setup.
rondoflow CLAUDE.md
Claude Code instructions for rondoflow/rondoflow, covering rondoflow, project overview, tech stack, development and setup.
claude-org-ja CLAUDE.md
Claude Code instructions for suisya-systems/claude-org-ja, covering secretary, 事実は出典付きで書く（citation-or-abstain）, 起動時, コミュニケーション and 輸送層（transport）両系 — 既定 renga / opt-in broker.