aimarket-mcp

aimarket-mcp

Enables web fetching, web search, and Metis verification for AI agents via a single MCP endpoint.

Category
访问服务器

README

<!-- aicom-mirror-notice -->

📖 Read-only mirror. aimarket-mcp is published from the canonical AI-Factory monorepo. Pull requests are not accepted — any commit pushed here is overwritten by scripts/mirror_satellites.sh on the next sync. 🐞 Found a bug or have a request? Please open an issue.

<!-- aicom-readme-badges --> <p align="center"> <a href="https://github.com/alexar76/aimarket-mcp/actions/workflows/ci.yml"><img src="https://img.shields.io/github/actions/workflow/status/alexar76/aimarket-mcp/ci.yml?branch=main&label=CI" alt="CI" /></a> <img src="https://img.shields.io/badge/MCP-gateway-6e40c9" alt="MCP gateway" /> <img src="https://img.shields.io/badge/tests-12_passing-brightgreen" alt="12 tests passing" /> <a href="docs/badges/coverage.svg"><img src="docs/badges/coverage.svg" alt="Test coverage" /></a> <a href="LICENSE"><img src="https://img.shields.io/badge/License-MIT-blue.svg" alt="License: MIT" /></a> </p> <!-- /aicom-readme-badges -->

aimarket-mcp — ecosystem MCP gateway

The alexar76 AICOM / AIMarket ecosystem's shared MCP gateway: one hardened server that gives every agent (Metis, ARGUS, …) the generic capabilities they were missing — web fetch, web search, and Metis verification — behind a single MCP endpoint.

It speaks MCP over Streamable-HTTP (JSON-RPC 2.0 POST, SSE-data:-framed responses, Mcp-Session-Id) — the exact protocol Metis's MCP client and ARGUS already talk, so no external SDK is required.

Tools

Tool What it does Hardening
web_fetch Fetch a URL, return its main text (readability-lite) SSRF-guarded (scheme allow-list, private-IP block, per-redirect re-validation, size cap); output sanitized + <untrusted>-wrapped
web_search Live DuckDuckGo search → top snippets output sanitized
metis_verify Run Metis's cognition + verification envelope on an input returns answer + machine-readable verify_score/verified

Why a gateway (not per-agent tools): generic capabilities are written once and every ecosystem agent gets them; the security core lives in one audited place. Ecosystem-specific capabilities already live in their own MCP servers (aimarket-oracle-gateway, aimarket-plugins); MNEMOSYNE-search and on-chain tools are the next additions here.

Security

  • SSRF: web_fetch validates the scheme (http/https only), blocks localhost/private/ reserved IPs, and re-validates on every redirect hop (vendored from Metis's audited security/ssrf.py).
  • Untrusted output: every web result is length-capped, has forged role markers stripped to a fixpoint, and is wrapped in <untrusted>…</untrusted> before it can reach a model.
  • Auth: optional bearer (AIMARKET_MCP_KEY); fail-closed in production (AIMARKET_MCP_PRODUCTION=1 requires a key).
  • Rate limit: per-key/IP token bucket (AIMARKET_MCP_RATE, default 120/min).

Run

pip install -e .
AIMARKET_MCP_KEY=sk-... AIMARKET_MCP_PRODUCTION=1 aimarket-mcp   # :9090
# or: docker compose up -d

Env: AIMARKET_MCP_PORT (9090) · AIMARKET_MCP_KEY · AIMARKET_MCP_PRODUCTION · AIMARKET_MCP_RATE · AIMARKET_METIS_URL (for metis_verify) · AIMARKET_METIS_KEY · AIMARKET_SEARCH_URL.

Consumers

  • Metis — enable the preset:
    enable_mcp_tools: true
    mcp_ecosystem_presets: [aimarket-web]
    
  • ARGUS — add the server to argus.config.json mcpServers (see that repo).

Test

pip install -e '.[dev]' && pytest -q

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选