mcp-client-kit

# mcp-client-kit **Write your MCP server wrappers once — from the live server. Keep them as real Python source you can diff, review, and pin.** `mcpgen` turns an MCP server reachable over **stdio or Streamable HTTP** into a Python module: one `async def` per tool, typed inputs from the start, and typed returns once you've probed the real responses. No live server needed to read it. `mcpgen` generates the Python wrapper layer for MCP code-mode architectures. Token savings occur when your agent or runtime executes these wrappers instead of loading the full MCP tool catalog into model context. `mcpgen` does not itself alter host tool registration and does not provide a sandbox. > **Package / repo / plugin:** `mcp-client-kit` · **CLI command:** `mcpgen` > > Two artifacts, one repo: a **CLI** (`mcpgen`) you run anywhere, and a **Claude Code plugin** (`generate-mcp-wrappers` skill) that drives it for the parts that need judgment. > This repo ships two artifacts: the `mcp-client-kit` Python package (PyPI; git tags `vX.Y.Z`; installs the `mcpgen` command) and the `mcp-client-kit` Claude Code plugin — which provides the `generate-mcp-wrappers` skill (git tags `plugin-vX.Y.Z`). --- ## The problem MCP tool schemas eat your context window before the agent does any work. Every tool definition costs **300–600 tokens** for its name, description, and JSON schema. That adds up fast: - The **GitHub MCP server alone** burns ~55,000 tokens across its 93 tools. - One developer measured **66,000 tokens consumed at conversation start** — a third of a 200k window, gone before the first query. - A SaaS server with 50+ endpoints can spend **30,000+ tokens just describing what it *could* do.** Anthropic's validated fix ("Code execution with MCP," Nov 2025): stop routing schemas through the model. Generate wrapper code and call the tools from code instead — an approach Anthropic measured shrinking one workflow from ~150,000 tokens to ~2,000 (**98.7%**), with independent benchmarks …