cookd CLAUDE.md

A project instruction file that tells the coding agent to consult a dependency map, recent session notes, and change-safety rules before editing code.

In plain words
What is it for?
Use it when working on cookd to check file dependencies, guardrails, and downstream effects before making changes.
Why use it?
It reduces the risk of breaking related parts of the project or overlooking established conventions.

Instructions file

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/codeclowns01/cookd/claude-md
Clone the repo
git clone --depth 1 https://github.com/codeclowns01/cookd
Per session 2,181 This file is loaded in full into every session.
When invoked 2,181 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 $0.02181 $0.02181
Opus 5 $0.01091 $0.01091
Sonnet 5 $0.00436 $0.00436
Haiku 4.5 $0.00218 $0.00218

Measured 2d ago against content hash dc29927a63db, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

cookd 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 2d 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 · 229 lines

How it starts

The opening of the file, as written. The whole thing — 229 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Claude Map Protocol

This project has a persistent intelligence map. Map location: ~/.claude-map/projects/cookd/ The map is the source of truth for dependencies, conventions, and change impact. When in doubt between the map and your own inference — trust the map, flag the discrepancy.


EVERY SESSION — Do this before anything else

Read these two files at the start of every session, without exception:

  • ~/.claude-map/projects/cookd/_sessions.md — what happened last session
  • ~/.claude-map/projects/cookd/_index.md — clusters, critical nodes, guardrail highlights

Then, for every file you plan to touch this session:

  • Read its map entry at ~/.claude-map/projects/cookd/map/[file].md

If a file you plan to touch has no map entry, create a stub for it before proceeding. Use the standard map template from the bootstrap. Mark it Confidence: unknown until you can fill it in.


BEFORE MAKING ANY CHANGE

For each file you are about to change:

Step 1 — Check the guardrails

Read the guardrail themes listed in that file's map entry. Load the relevant fragment from ~/.claude-map/projects/cookd/guardrails/[theme].md. If the change you are about to make would violate a guardrail, stop and flag it to the user before proceeding. Do not silently work around a guardrail.

Step 2 — Determine if this is an interface change or implementation change

Read the "Interface vs Implementation" section of the file's map entry.

  • Implementation change (internal logic only, nothing exported changes): proceed. No downstream review needed.
  • Interface change (exported functions, API shape, types, env vars, response format): go to Step 3.

Step 3 — Walk the downstream list (interface changes only)

Read the "Downstream Dependents" list in the file's map entry. For each downstream file listed:

  • Check whether the change affects that file's dependency on this one
  • If yes: flag it to the user as a required co-change or review
  • If no: note it as reviewed and unaffected

Read the full file on GitHub · 229 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. 2d ago First seen · 229 lines · 2,181 tokens per session scan A dc29927a63db

Subscribe to this mod's changes

cookd CLAUDE.md is an instructions file published in the GitHub repository codeclowns01/cookd (5 stars, last pushed 12d ago), licensed MIT. It adds 2,181 tokens to every session, about $0.0109 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.