Clew CLAUDE.md

Clew CLAUDE.md is an instructions file for Claude Code from miuuyy/Clew. It costs 606 tokens per session, scanned A, original, MIT.

Repository instructions for Clew, a local-first learning workspace built around a graph of knowledge and an AI agent. They tell Claude Code what to read first and which design boundaries to preserve.

In plain words
What is it for?
Use them when modifying Clew’s graph workspace, agent behavior, snapshots, quizzes, imports, exports, or user interface.
Why use it?
They give the coding agent project context and guardrails, reducing the risk of changes that silently apply AI suggestions, break reversibility, or move the product away from its graph-centered design.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md. Also seen: reads .claude/ paths; mentions CLAUDE.md; mentions Claude Code.

Install

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.

agentmods
npx agentmods add instructions/miuuyy/clew/claude-md
Clone the repo
git clone --depth 1 https://github.com/miuuyy/Clew

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.

agentmods badge for Clew CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/miuuyy/clew/claude-md.svg)](https://agentmods.dev/instructions/miuuyy/clew/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/miuuyy/clew/claude-md"><img src="https://agentmods.dev/badge/instructions/miuuyy/clew/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 606 This file is loaded in full into every session.
When invoked 606 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00606 $0.00606
Opus 5 $0.00303 $0.00303
Sonnet 5 $0.00121 $0.00121
Haiku 4.5 $0.00061 $0.00061

Measured 6d ago against content hash a00c38d096ba, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

Clew 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 6d 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 · 93 lines

How it starts

The opening of the file, as written. The whole thing — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

This file provides repository-specific guidance to Claude Code when working in Clew.

Read first

Before making changes, read:

Product overview

Clew is a graph-first learning workspace with an AI agent layer.

If Obsidian is where you collect notes, Clew is where you operate on a learning graph.

The public main branch is the local developer edition. It keeps:

  • graph workspace
  • provider seam
  • proposals
  • snapshots and rollback
  • quizzes and closure logic
  • import/export

It does not try to mirror every hosted product surface.

Core boundaries

Never weaken these:

  1. The graph is the center of truth.
  2. AI suggests changes instead of silently applying them.
  3. Accepted graph changes stay reversible through snapshots.
  4. The UI stays graph-first instead of dashboard-first.
  5. The public local edition stays honest about being local-first.

Engineering rules

  • Prefer explicit contracts over heuristics.
  • Prefer fail-closed over hidden fallback.
  • Do not add semantic keyword routers for agent behavior.
  • Keep model behavior honest through role, context, and schema.
  • Do not add dead surfaces “just in case”.

Frontend quick map

  • frontend/src/App.tsx: app shell and state wiring
  • frontend/src/components/WorkspaceShell.tsx: sidebar and workspace chrome
  • frontend/src/components/GraphCanvas.tsx: graph rendering and interaction
  • frontend/src/components/SettingsModal.tsx: provider, memory, debug, and UX controls
  • frontend/src/lib/: API, graph helpers, copy, contracts, debug logging

Backend quick map

  • backend/app/api/routes.py: HTTP transport
  • backend/app/services/repository.py: SQLite persistence, snapshots, config
  • backend/app/services/chat_orchestrator.py: answer/quiz/proposal action choice
  • backend/app/services/proposal_planner.py: proposal generation and validation bridge
  • backend/app/llm/: providers, catalog, prompts, contracts, schemas

Read the full file on GitHub · 93 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.

  1. 6d ago First seen · 93 lines · 606 tokens per session scan A a00c38d096ba

Subscribe to this mod's changes

Clew CLAUDE.md is an instructions file published in the GitHub repository miuuyy/Clew (165 stars, last pushed today), licensed MIT. It adds 606 tokens to every session, about $0.0030 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.