genesis-mcp: Instructions file for Claude Code

CLAUDE.md

genesis-mcp CLAUDE.md is an instructions file for Claude Code from adeoluwaadesina/genesis-mcp. It costs 1,041 tokens per session, scanned A, original, MIT.

Repository instructions for Genesis MCP, a server that can add tools to its in-memory registry while it is running. The host AI writes the generated Python code; Genesis only builds the prompt and registers the resulting tool.

In plain words
What is it for?
They guide installation, server startup, testing, and changes to dynamic tool creation, prompt assembly, registration, and hot-loading.
Why use it?
They clarify how tool creation actually works and prevent an agent from assuming the server calls an AI model or needs a restart for every new tool.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md. Also seen: mentions CLAUDE.md; mentions Claude Code.

This is adeoluwaadesina/genesis-mcp's own configuration. It tells Claude Code how to work on genesis-mcp 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 genesis-mcp configures →

Reuse

Borrowing it

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

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 genesis-mcp CLAUDE.md

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

Your own site · 80×15
<a href="https://agentmods.dev/instructions/adeoluwaadesina/genesis-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/adeoluwaadesina/genesis-mcp/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 1,041 This file is loaded in full into every session.
When invoked 1,041 The same file — it is already loaded in full.
Security scan A 1 finding. 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 $0.01041 $0.01041
Opus 5.5 $0.00416 $0.00416
Sonnet 5.5 $0.00208 $0.00208
Haiku 4.5 $0.00104 $0.00104

Measured 2mo ago against content hash 12b5295097ee, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

genesis-mcp CLAUDE.md scanned grade A with 1 finding 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 2mo 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.

Runs shell commandslowCapability

Expected in a hook, worth knowing in a rule or an instructions file.

3. Forbidden calls (`eval`, `exec`, `compile`, `__import__`, `os.system`, `subprocess.*`, etc.)
CLAUDE.md · 87 lines

How it starts

The opening of the file, as written. The whole thing — 87 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.

Commands

# Install (editable, with dev deps)
pip install -e ".[dev]"

# Run the MCP server
python -m genesis.server

# Run all tests
pytest tests/

# Run a single test file
pytest tests/test_validator.py -v

Architecture

Genesis MCP is a self-extending MCP server. The key design: it uses the low-level mcp.Server class (not FastMCP) so that list_tools and call_tool are implemented as dynamic handlers that read from the in-memory registry at call time. This is what makes hot-loading possible without a server restart.

Important: Genesis does not call any LLM API server-side. Code generation runs on the host LLM (Claude Desktop, Cursor, etc.). Genesis only assembles a prompt and accepts the resulting code back through register_tool.

Data flow for tool creation

User asks Claude to make a tool
  → host LLM calls meta_tools.create_tool(description)
      → generator.build_generation_prompt()        # returns a prompt STRING; no API call
  → host LLM reads the prompt, generates Python code locally
  → host LLM calls meta_tools.register_tool(code=...)
      → validator.validate()                        # 8 checks, fail-fast
      → write generated_tools/<name>.py
      → loader.load_tool_file()                     # importlib dynamic load
      → registry.register()                         # in-memory + tools_manifest.json

The trust boundary is register_tool: any string that passes validator.validate() is loaded into the live server process.

Key files

File Responsibility
genesis/server.py Entry point. Low-level MCP server with dynamic dispatch.
genesis/meta_tools.py The 6 built-in meta-tools (create_tool, register_tool, list_tools, delete_tool, update_tool, describe_tool). META_TOOL_SCHEMAS is read by server at boot.
genesis/generator.py Builds the natural-language prompt that the host LLM consumes. No API calls.
genesis/validator.py AST-based checks: syntax, imports, forbidden calls, forbidden names (reflection), structure, schema, uniqueness, side effects.
genesis/registry.py In-memory {name: ToolEntry} dict. Persists to tools_manifest.json.
genesis/loader.py importlib.util dynamic loading. Raises LoadError on all failure modes.
genesis/config.py Singleton config loaded from config.yaml. Also calls load_dotenv for future-compat, but no subsystem reads env vars in v0.1.

Read the full file on GitHub · 87 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. 2mo ago First seen · 87 lines · 1,041 tokens per session scan A 12b5295097ee

Subscribe to this mod's changes

genesis-mcp CLAUDE.md is an instructions file published in the GitHub repository adeoluwaadesina/genesis-mcp (2 stars, last pushed 5mo ago), licensed MIT. It adds 1,041 tokens to every session, about $0.0042 per session on Opus 5.5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

genesis-mcp CLAUDE.md compared with instructions from other repositories

genesis-mcp CLAUDE.md next to 6 instructions from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
genesis-mcp CLAUDE.mdadeoluwaadesina/genesis-mcpInstructions file25mo agoMITA1,041Claude 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