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/yeison-liscano/http_mcp/claude-mdgit clone --depth 1 https://github.com/yeison-liscano/http_mcpWrote 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.
[](https://agentmods.dev/instructions/yeison-liscano/http_mcp/claude-md)<a href="https://agentmods.dev/instructions/yeison-liscano/http_mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/yeison-liscano/http_mcp/claude-md.svg" alt="Measured on agentmods" height="20"></a>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 5.1 | $0.01772 | $0.01772 |
| Opus 5 | $0.00886 | $0.00886 |
| Sonnet 5 | $0.00354 | $0.00354 |
| Haiku 4.5 | $0.00177 | $0.00177 |
Grade A, and why
http_mcp 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 5d 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.
How it starts
The opening of the file, as written. The whole thing — 156 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
http-mcp is a lightweight Python library implementing the Model Context
Protocol (MCP) over HTTP and STDIO. It exposes Python functions as tools and
prompts via JSON-RPC, designed for use with Starlette/FastAPI applications.
Published on PyPI as http-mcp.
Common Commands
# Install dependencies
uv sync
# Run all tests with coverage (90% minimum required)
pytest --cov-report term-missing --cov=src --cov-fail-under=90 tests/
# Run a single test file
pytest tests/test_tools.py
# Run a single test
pytest tests/test_tools.py::test_name -v
# Type checking (strict mode)
mypy .
# Lint (all ruff rules enabled)
ruff check
# Format
ruff format
# Format markdown files (never `mdformat .` — it rewrites the YAML
# frontmatter in .claude/ into a heading and breaks the agent definitions)
mdformat README.md docs/ --wrap 80 --exclude .claude/
# Full pre-push check (what CI runs)
ruff check && mypy . && pytest --cov-report term-missing --cov=src --cov-fail-under=90 tests/ && mdformat README.md docs/ --wrap 80 --exclude .claude/
Architecture
Source Layout
Source is in src/http_mcp/ and src/auth_mcp/ with hatchling as build
backend. The package uses uv for dependency management. Both packages ship in
the same wheel (http-mcp); auth_mcp is available via
pip install http-mcp[auth].
Core Components
server.py—MCPServerclass: central entry point. Manages tools/prompts, handles JSON-RPC dispatch, supports both HTTP and STDIO transports. Integrates with Starlette's lifespan for state management.server_interface.py— Abstract base class (ServerInterface) thatMCPServerimplements.types/— Public API types exported viahttp_mcp.types:Tool— Wraps a function with Pydantic input/output schemas. Supports sync/async, optionalArguments[TInputs]parameter, error handling, and authorization scopes.Prompt— Interactive templates returningtuple[PromptMessage, ...].Arguments[T]— Generic container providinginputs,request, andget_state_key()for accessing lifespan state.NoArguments— Empty Pydantic model for tools/prompts without parameters.
exceptions.py— Exception hierarchy:BaseError→ProtocolError,ServerError(withToolNotFoundError,ToolInvocationError,PromptNotFoundError,PromptInvocationError),ArgumentsError. Error codes live in_json_rcp_types/errors.py;-32002was retired by the 2026-07-28 revision and must not be reintroduced.
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.
- 5d ago First seen · 156 lines · 1,772 tokens per session scan A 77f293817fcc
http_mcp CLAUDE.md is an instructions file published in the GitHub repository yeison-liscano/http_mcp (0 stars, last pushed 9d ago), licensed MIT. It adds 1,772 tokens to every session, about $0.0089 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
monkey AGENTS.md
AGENTS.md instructions for monkey/monkey, covering agents, repository map, main runtime flow, build and verification and commit style used in this repository.
KtorMonitor AGENTS.md
AGENTS.md instructions for CosminMihuMDC/KtorMonitor, covering agents.md — ktormonitor, 1. project overview, 2. module structure, source-set layout (core) and 3. build, test, run.
voleeo-api AGENTS.md
AGENTS.md instructions for voleeo/voleeo-api, covering agents.md, product vision, model routing (delegated sub-work only), tool discipline and code quality.
httpstat AGENTS.md
Instructions for reorx/httpstat, covering httpstat 项目规划, 1) 项目现状, 已完成的现代化工作, 2) 剩余技术债 and a. 测试可靠性不足(高优先级).
casbin-gateway CLAUDE.md
Claude Code instructions for apache/casbin-gateway, covering claude.md, code style and chinese translations.
igniter-js GEMINI.md
Gemini CLI instructions for felipebarcelospro/igniter-js, covering 1. identity and profile, 2. about the igniter.js monorepo, 3. personality and communication, 4. lia's core responsibilities (the 4 pillars) and 5. technical guidelines and methodology.