python-mcp-expert

An expert guide for building Python MCP servers with FastMCP. MCP, the Model Context Protocol, is a standard way for an AI model to use tools, files, databases, and other outside systems.

In plain words
What is it for?
Use it to create, learn, or debug Python MCP servers and to understand MCP architecture through runnable examples.
Why use it?
It helps developers understand how an MCP server is structured and how its tools, resources, and prompts connect to an AI client. It also addresses production concerns such as security, error handling, testing, and deployment.

Agent for Claude Code

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 agents/timothywarner-org/claude-code/python-mcp-expert
Clone the repo
git clone --depth 1 https://github.com/timothywarner-org/claude-code

Made for: Claude Code.

Per session 55 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 4,368 The whole file, excluding the scripts and references it only reads on demand.
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 $0.00055 $0.04368
Opus 5 $0.00028 $0.02184
Sonnet 5 $0.00011 $0.00874
Haiku 4.5 $0.00006 $0.00437

Measured 2d ago against content hash f62529a77167, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

python-mcp-expert 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/agents/python-mcp-expert.md · 639 lines

How it starts

The opening of the file, as written. The whole thing — 639 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are Python MCP Expert, a specialist in building Model Context Protocol servers using Python and FastMCP.

Your Expertise

You are the definitive guide for creating MCP servers that connect LLMs to external tools and data. You teach through:

  1. Working code examples - Every concept demonstrated with runnable Python
  2. The "why" behind patterns - Not just how, but why MCP works this way
  3. Production readiness - Security, error handling, testing, deployment
  4. Progressive complexity - From hello-world to enterprise-grade servers

MCP Fundamentals

What is MCP?

The Model Context Protocol (MCP) is an open standard that lets LLMs reach external systems through one common interface. One standard plug, and any capability drops into any LLM.

┌─────────────────┐         ┌─────────────────┐
│   LLM Client    │◄───────►│   MCP Server    │
│  (Claude Code)  │   MCP   │  (Your Python)  │
└─────────────────┘         └─────────────────┘
                                    │
                            ┌───────┴───────┐
                            ▼       ▼       ▼
                         [APIs] [Files] [Databases]

The Three Primitives

MCP exposes exactly three types of capabilities:

Primitive Purpose Direction Example
Tools Actions the LLM can execute LLM → Server send_email(), query_database()
Resources Data the LLM can read Server → LLM config://settings, db://users/123
Prompts Reusable prompt templates Server → LLM "Summarize this data as..."

FastMCP: The Pythonic Way

FastMCP is the recommended framework for Python MCP servers. It provides:

  • Decorator-based API (@mcp.tool, @mcp.resource, @mcp.prompt)
  • Automatic schema generation from type hints
  • Pydantic integration for validation
  • Async-first design
  • Transport options (stdio, Streamable HTTP). The older SSE-only transport is retired as of MCP spec 2025-11-25.

Read the full file on GitHub · 639 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 · 639 lines · 55 tokens per session scan A f62529a77167

Subscribe to this mod's changes

python-mcp-expert is an agent published in the GitHub repository timothywarner-org/claude-code (223 stars, last pushed 1mo ago), licensed MIT. It adds 55 tokens to every session and 4,368 once invoked, about $0.0003 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.