What is claude_code CLAUDE.md?

A project guide for Claude Code, an AI coding tool, working on an Elixir software library that communicates with the Claude Code command-line program.

Installing dependencies, running formatting and static checks, executing tests, checking types, and managing the bundled command-line program.

It puts the project's setup commands, quality checks, tests, and development conventions in one place.

claude_code CLAUDE.md — Instructions file by guess

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/guess/claude_code/claude-md

Clone the repo

git clone --depth 1 https://github.com/guess/claude_code

Per session 3,617 This file is loaded in full into every session.

When invoked 3,617 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.

Model	Per session	Once invoked
Fable 5	$0.03617	$0.03617
Opus 5	$0.01809	$0.01809
Sonnet 5	$0.00723	$0.00723
Haiku 4.5	$0.00362	$0.00362

Measured 2d ago against content hash d813b93668d1, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

claude_code 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 · 356 lines

How it starts

The opening of the file, as written. The whole thing — 356 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.

Project Overview

This is the ClaudeCode Elixir SDK - an idiomatic Elixir interface to the Claude Code CLI. The SDK spawns the claude command as a subprocess and communicates via streaming JSON over stdout.

Common Development Commands

Quality Checks

mix quality          # Run all code quality checks (compile, format, credo, dialyzer)
mix format           # Format code with Styler
mix credo --strict   # Run Credo analysis
mix dialyzer         # Run Dialyzer type checking

Testing

mix test                        # Run all tests
mix test test/path/to_test.exs # Run specific test file
mix test.all                    # Run tests with coverage report
mix coveralls.html              # Generate HTML coverage report

# Distributed tests (Adapter.Node) require a named BEAM node:
elixir --sname test -S mix test --include distributed

Development

mix deps.get              # Install dependencies
mix claude_code.install   # Install CLI binary to priv/bin/
mix claude_code.uninstall # Remove bundled CLI binary
mix claude_code.path      # Print resolved CLI binary path
iex -S mix                # Start interactive shell with project loaded
mix docs                  # Generate documentation

Architecture

The SDK is organized in three layers:

Adapter-agnostic — Session, Stream, Options (validation), Types, Message/Content structs
CLI protocol — CLI.Command (flags), CLI.Input (stdin), CLI.Parser (JSON parsing)
Adapters — Adapter.Port (local CLI via Port), Adapter.Node (distributed via BEAM), Adapter.Test (in-memory stubs)

Key modules:

ClaudeCode.Session - Public API for session operations (lifecycle, runtime config, MCP, introspection, execute)
ClaudeCode.Session.Server - GenServer that manages the adapter subprocess lifecycle
ClaudeCode.Adapter.Port - Port-based adapter that spawns the CLI as a local subprocess
ClaudeCode.Adapter.Node - Distributed adapter that runs Adapter.Port on a remote BEAM node via Erlang distribution
ClaudeCode.CLI - Thin facade: binary resolution + command building (delegates to Resolver + Command)
ClaudeCode.CLI.Command - Converts Elixir options to CLI flags and builds argument lists
ClaudeCode.CLI.Parser - Parses newline-delimited JSON from CLI output into structs
ClaudeCode.Adapter.Port.Installer - CLI binary download and version management
ClaudeCode.Plugin - Plugin management (list, install, uninstall, enable, disable, update)
ClaudeCode.Plugin.Marketplace - Marketplace management (list, add, remove, update)

Read the full file on GitHub · 356 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.

2d ago First seen · 356 lines · 3,617 tokens per session scan A d813b93668d1

Subscribe to this mod's changes

claude_code CLAUDE.md is an instructions file published in the GitHub repository guess/claude_code (96 stars, last pushed 2mo ago), licensed MIT. It adds 3,617 tokens to every session, about $0.0181 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.

Languages Testing ai anthropic beam claude elixir genserver

loomkin AGENTS.md

Instructions for pass-agent/loomkin, covering loomkin - ai instructions, monorepo layout, authentication, routes that require authentication and routes that work with or without authentication.

pass-agent/loomkin · 686 tokens

loomkin copilot-instructions.md

Instructions for pass-agent/loomkin: Read and follow CONTRIBUTING.md for all conventions, including commit message format, build commands, and code style.

pass-agent/loomkin · 91 tokens

loomkin CLAUDE.md

Instructions for pass-agent/loomkin, a project described as: What if AI agents could form teams as fluidly as humans? Spawn specialists in milliseconds, share discoveries in real-time, review each other's work, and remember everything across sessions. Watch it all unfold in a live mission control UI. Open source.

pass-agent/loomkin · 3 tokens

claude_code CLAUDE.md