agentmods.dev

medusa: Instructions file for Claude Code

CLAUDE.md

medusa CLAUDE.md is an instructions file for Claude Code from Skowt/medusa. It costs 25,456 tokens per session, scanned D, original, MIT.

A set of instructions for coding agents working on the Medusa Go repository. It explains the project’s architecture, common commands, and the checks required before a Go change is considered complete.

In plain words
What is it for?
Use it when making changes to Medusa, especially Go code under internal/app. It guides building, formatting, targeted tests, race-detector runs, linting, and final CI-like verification.
Why use it?
It gives contributors a shared definition of done and reduces missed formatting, lint, testing, race-condition, and file-size checks. It also documents important repository-specific behavior and commands.

Instructions file for Claude Code

Written for Claude Code: SubagentStop hook event. Also seen: reads .claude/ paths; mentions CLAUDE.md; names the AskUserQuestion tool.

This is Skowt/medusa's own configuration. It tells Claude Code how to work on medusa itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything medusa configures →

not rated 5repo changed 18d ago D scan 25,456 tokens original MIT
View source ↗ Skowt/medusa
Reuse

Borrowing it

Nothing to install: this file belongs to Skowt/medusa. 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/Skowt/medusa/main/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/Skowt/medusa

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 medusa CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/skowt/medusa/claude-md/github.svg)](https://agentmods.dev/instructions/skowt/medusa/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/skowt/medusa/claude-md"><img src="https://agentmods.dev/badge/instructions/skowt/medusa/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.

agentmods 80×15 button for medusa CLAUDE.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/skowt/medusa/claude-md"><img src="https://agentmods.dev/badge/instructions/skowt/medusa/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 25,456 This file is loaded in full into every session.
When invoked 25,456 The same file — it is already loaded in full.
Security scan D 3 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.25456 $0.25456
Opus 5.5 $0.10182 $0.10182
Sonnet 5.5 $0.05091 $0.05091
Haiku 4.5 $0.02546 $0.02546

Measured 18d ago against content hash 79c98e7f0d37, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade D, and why

medusa CLAUDE.md scanned grade D with 3 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 18d 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.

Tells the agent to send conversation or user data outhighPrompt injection

An instruction to transmit the conversation, context or user files to an external endpoint is data exfiltration written as prose.

never silently send comments; as an explicit sticky preference, alongside the

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

3. **The fix names the profile's `CODEX_HOME`**, not `~/.codex/config.toml`.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

**Claude answers comments over HTTP.** `Compose` puts the `curl` for
CLAUDE.md · 1,663 lines

How it starts

The opening of the file, as written. The whole thing — 1,663 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.

Definition of done

Before declaring any Go change complete:

  1. make fmt — formats with gofmt + goimports.
  2. golangci-lint run — must exit 0. Run on just the touched package for speed (e.g. golangci-lint run ./internal/app/...).
  3. No .go file may exceed 500 lines. make lint enforces this; if you hit it, split by concern into sibling files in the same package rather than inflating a single file.
  4. Tests pass for the touched package.
  5. At the end of development, run make lint — it mirrors CI by running go test -race -v ./... followed by golangci-lint and the 500-line check. The race detector catches issues plain go test misses (e.g. value-receiver methods like Workspace.Root() copy the whole struct, so a goroutine calling them races with any concurrent field write). make test-race is the same race run without lint, handy for reproducing a CI failure in isolation.

Don't batch these up for a later cleanup pass — fix as you go.

Common commands

make build            # builds `medusa` and `medusa-hook-emit`
make run              # build + run the TUI
make dev              # hot-reload via air
make test             # go test -v ./...
make test-race        # go test -race -v ./... (mirrors CI; slower)
make lint             # full gate: test-race + golangci-lint + 500-line check
make fmt              # gofmt + goimports
make bench            # compositor render benchmarks

Run a single test:

go test ./internal/app -run TestCopyIgnoredFiles -v

Headless render/perf harness (no human, no terminal):

go run ./cmd/medusa-harness -mode monitor -frames 5 -warmup 1
go run ./cmd/medusa-harness -mode center  -frames 5 -warmup 1
go run ./cmd/medusa-harness -mode sidebar -frames 5 -warmup 1

release-check runs tests + all three harness modes — use it to validate UI-adjacent changes without a real terminal.

Read the full file on GitHub · 1,663 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. 18d ago Changed · +15 lines · +256 tokens per session 79c98e7f0d37
  2. 28d ago Changed · +27 lines · +421 tokens per session 9d7386668a30
  3. 29d ago Changed · +54 lines · +928 tokens per session 89c8e56bee1c
  4. 1mo ago Changed · +6 lines · +113 tokens per session a6ef5a8f6075
  5. 1mo ago Changed · +3 lines · +58 tokens per session ff1f5e6f4731
  6. 1mo ago Changed · +887 lines · +13,633 tokens per session scan A → D 7e1e0f328aa6
  7. 2mo ago Changed · +380 lines · +5,564 tokens per session 2a8ec4f68cef
  8. 2mo ago First seen · 291 lines · 4,483 tokens per session scan A 6aaf080a2c6c

Subscribe to this mod's changes

medusa CLAUDE.md is an instructions file published in the GitHub repository Skowt/medusa (5 stars, last pushed 10d ago), licensed MIT. It adds 25,456 tokens to every session, about $0.1018 per session on Opus 5.5. A static security scan graded it D with 3 findings (tells the agent to send conversation or user data out, reads agent configuration directories, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

medusa CLAUDE.md compared with instructions from other repositories

medusa CLAUDE.md next to 6 instructions from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
medusa CLAUDE.mdSkowt/medusaInstructions file510d agoMITD25,456Claude Code
vscode buildNext.instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA9,719GitHub Copilot
next.js AGENTS.mdvercel/next.jsInstructions file143kyesterdayMITA7,461Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchainInstructions file149kyesterdayMITA4,680Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA5,516GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA3,093GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harnessInstructions file274k2d agoMITA4,313Codex, OpenCode