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/npatten/clew/agents-mdgit clone --depth 1 https://github.com/npatten/ClewWhat 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 | $0.01722 | $0.01722 |
| Opus 5 | $0.00861 | $0.00861 |
| Sonnet 5 | $0.00344 | $0.00344 |
| Haiku 4.5 | $0.00172 | $0.00172 |
Grade A, and why
Clew AGENTS.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 yesterday.
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 — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Guidelines
These are the guidelines and rules for agents working on the Clew project. In Greek myth, Ariadne gave Theseus the "Clew" (a ball of thread) to navigate the labyrinth so he could find his way back out. It represents the lightweight tracking of Tasks and Increments ensuring the codebase remains stable and you never lose your way.
Resources
- Full project spec:
clew-spec.md - All open work items run:
clew list
Software development
You are the principal engineer — guide the process, use sub-agents in parallel where appropriate, avoid unnecessary complexity and tech debt.
- Ask before removing functionality or code that appears intentional
- Do not preserve backward compatibility unless the user explicitly asks
- Prefer functional style, simple composable components
- Raise larger refactors with the user before starting
- Conserve context: scout sub-agents for whole-file reads; parallel sub-agents for independent tasks
Design discipline
- Simple ≠ easy. Familiar is not simple. Before choosing a construct, ask: does it complect (braid together) concerns that could be independent? Can the next reader reason about one piece without loading the others? "I already know this pattern" is not an argument.
- Modularity is not simplicity. Splitting tangled code across files leaves it tangled. Watch for hidden coupling: shared mutable state, implicit ordering, call-site assumptions about internals.
- Name the downside. Any tool, pattern, or abstraction you propose — state its cost alongside its benefit. "Has benefit X" without "and costs Y" is incomplete analysis.
Code organization
- Modern module style —
core.rs+core/directory, notcore/mod.rs. core/is pure logic (no I/O);storage/is the I/O seam;commands/orchestrates. Keep the seam clean — pure logic stays unit-testable without tempdirs.lib.rs+main.rssplit —main.rsstays thin; integration tests import the library.
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.
- yesterday First seen · 161 lines · 1,722 tokens per session scan A 0326f6c9108e
Clew AGENTS.md is an instructions file published in the GitHub repository npatten/Clew (2 stars, last pushed 3mo ago), licensed MIT. It adds 1,722 tokens to every session, about $0.0086 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
chainlink CLAUDE.md
Instructions for dollspace-gay/chainlink, covering chainlink issue tracker, commands, issues, organization and sessions.
orbit CLAUDE.md
Instructions for Noveum/orbit, covering orbit, hard rules, bun is the toolchain, not the runtime, layout and commands.
orbit AGENTS.md
Instructions for Noveum/orbit, covering agent instructions, the short version, rules that will fail the build and before you report a task finished.
lineark CLAUDE.md
Instructions for flipbit03/lineark, covering claude.md — lineark, what is this, workspace layout, key commands and writing online tests: the three pillars.
dirge CLAUDE.md
Claude Code instructions for dirge-code/dirge, covering project instructions for ai agents, beads issue tracker, quick reference, rules and session completion.
moo-tasks AGENTS.md
Instructions for dizlexic/moo-tasks, covering agents.md, 1. what is the moo tasks board?, 2. connecting your agent, 3. available mcp tools and 4. recommended workflow for agents.