amurshak/congressmcp

开发工具

本页由 AI 翻译，技术细节请以原文为准。查看原文

36 stars 11 forks 质量 79 趋势 20NOASSERTION

Repository Install

amurshak/CongressMCP 是从 https://github.com/amurshak/congressmcp. 导入的 MCP 服务器候选人。

概览

amurshak/CongressMCP 是从 https://github.com/amurshak/congressmcp. 导入的 MCP 服务器候选人。

概述（AI 翻译）

任何MCP客户端的实时美国国会数据——Claude Code、ChatGPT、Copilot、Codex、Cursor、OpenCode、Gemini CLI、Grok Build等。

法案、完整法案文本、投票、成员、委员会、听证会、提名和国会记录——通过模型上下文协议以自然语言查询。在你的机器上本地运行，支持免费 Congress.gov 和GovInfo API。无账户，无托管服务，无遥测。

快速启动

1.获取免费 Congress.gov API密钥

注册地址为api.congress.gov/sign-up——只需30秒。同样的密钥也适用于GovInfo（账单全文）。

2.安装 `uv`

CongressMCP 在 PyPI 上发布，并与 uvx 一起发布，该版本带 UV：

（brew install uv、winget install astral-sh.uv和pipx install uv也可用。）喜欢点数？pip install congressmcp给你一个congressmcp命令，你可以用来替代下面uvx congressmcp]。

3.连接你的客户

每个客户端都需要相同的三个事实：command uvx，args ["congressmcp"]，env CONGRESS_API_KEY。

本页由 AI 翻译，技术细节请以原文为准。

README

以下为原文 README 节选，完整内容请查看原文。

CongressMCP

Live U.S. Congressional data for any MCP client — Claude Code, ChatGPT, Copilot, Codex, Cursor, OpenCode, Gemini CLI, Grok Build, and more.

Bills, full bill text, votes, members, committees, hearings, nominations, and the Congressional Record — queried in natural language through the Model Context Protocol. Runs locally on your machine against the free Congress.gov and GovInfo APIs. No account, no hosted service, no telemetry.

Quick Start

1. Get a free Congress.gov API key

2. Install `uv`

CongressMCP is published on PyPI and launched with uvx, which ships with uv:

# macOS / Linux
curl -LsSf https://astral.sh/uv/install.sh | sh

# Windows (PowerShell)
powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

(brew install uv, winget install astral-sh.uv, and pipx install uv also work.) Prefer pip? pip install congressmcp gives you a congressmcp command you can use in place of uvx congressmcp below.

3. Connect your client

Every client needs the same three facts: command uvx, args ["congressmcp"], env CONGRESS_API_KEY. Clients are listed roughly by how many professional developers use them today (JetBrains Developer Ecosystem survey, mid-2026, then Pragmatic Engineer’s 2026 tooling survey), so the one you want is probably near the top:

Client	Where it’s configured	Notes
Claude Code	`claude mcp add …` or `.mcp.json`
ChatGPT	Developer mode → connector URL	remote only — needs HTTP mode
VS Code / GitHub Copilot	`.vscode/mcp.json`	uses `servers` + `inputs`
OpenAI Codex CLI	`codex mcp add …` or `~/.codex/config.toml`	TOML
Cursor	`~/.cursor/mcp.json` or `.cursor/mcp.json`
JetBrains AI Assistant / Junie	Settings → AI Assistant → MCP	paste the Claude Desktop JSON
OpenCode	`opencode.json` → `mcp`	`command` is a single array
Gemini CLI	`gemini mcp add …` or `~/.gemini/settings.json`
Claude.ai	Connectors → custom connector URL	remote only — needs HTTP mode
Claude Desktop	`claude_desktop_config.json`
Windsurf	`~/.codeium/windsurf/mcp_config.json`
Zed	`settings.json` → `context_servers`
Cline / Roo Code	MCP settings panel → edit JSON
Goose	`goose configure` or `~/.config/goose/config.yaml`	YAML

在 GitHub 查看完整 README

安装

uvx congressmcp

新手？先看 mcpskill.org 教程 →

配置

{
  "mcpServers": {
    "congressmcp": {
      "type": "stdio",
      "command": "uvx",
      "args": ["congressmcp"],
      "env": { "CONGRESS_API_KEY": "${CONGRESS_API_KEY}" }
    }
  }
}