AIMarket MCP Gateway

io.github.alexar76v0.3.0更新於 Oct 1, 2026

Stdio + HTTP MCP gateway: SSRF-hardened web_fetch, web_search, metis_verify, market_search.

已驗證Streamable HTTP可網頁執行Web Search & ScrapingAI & MLBusiness & Commerce

概覽

AI 產生的概覽

一個 MCP 閘道，提供強化過的網頁抓取、網頁搜尋、Metis 驗證，以及 AIMarket 能力探索與試用呼叫工具。

功能
它在同一個 MCP 伺服器裡提供五個工具：web_fetch 抓取 URL 並回傳正文，具備 SSRF 防護並將輸出標記為不可信；web_search 回傳 DuckDuckGo 即時摘要；metis_verify 回傳 Metis 答案與 verify_score、verified 判定；market_search 在 AIMarket 中樞上探索計價能力；market_invoke 在中樞的免費試用層執行一項能力並回傳簽章收據 nonce。它可透過 stdio 供桌面用戶端使用，也能自架為 9090 埠的 Streamable-HTTP 服務。
適用情境
當助理需要網頁檢索並附帶驗證步驟時適用，也適合在不安裝各代理獨立工具的情況下瀏覽與試用 AIMarket 中樞能力。託管端點只涵蓋市場類工具，因此若還需要 web_fetch、web_search 與 metis_verify，或想把市場工具指向自己的中樞，就應安裝本套件。
執行需求
需要 Python 與 aimarket-mcp 套件（uvx 或 pip），或透過 Docker Compose 自架 HTTP 部署。選用環境變數：AIMARKET_METIS_URL、AIMARKET_METIS_KEY、AIMARKET_SEARCH_URL、AIMARKET_HUB_URL、AIMARKET_SANDBOX_VISITOR。HTTP 模式另需 AIMARKET_MCP_KEY、AIMARKET_MCP_PRODUCTION、AIMARKET_MCP_RATE、AIMARKET_MCP_PORT，用戶端需送出 Authorization bearer 標頭。需要連線至所設端點的網路。
安裝前請注意
AIMARKET_METIS_KEY 與 HTTP 的 Authorization bearer 屬於機密，應以環境變數或標頭提供，不要提交進版控。market_invoke 會消耗免費試用額度，額度用盡時會原樣回報中樞的 402；鏈上付費託管是另一套流程。web_fetch 與 web_search 會連外至外部網站，抓取或搜尋到的內容以不可信形式回傳。HTTP 模式下應設定 AIMARKET_MCP_PRODUCTION=1，讓缺少 AIMARKET_MCP_KEY 時以失敗關閉處理。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 AIMarket MCP Gateway，將其新增到工作區。
  2. 為需要使用其工具的對話啟用該服務。

Web executable，透過 Streamable HTTP。 遠端服務在工作區中設定後即可從網頁執行環境執行。

其他 MCP 客戶端

把它新增到你客戶端的 mcpServers 設定中。

{
  "mcpServers": {
    "aimarket-mcp": {
      "type": "http",
      "url": "https://{aimarket_host}/mcp"
    }
  }
}

README

📖 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.

aimarket-mcp — ecosystem MCP gateway

[CI] [aimarket-mcp MCP server] [MCP gateway] [stdio + HTTP] [14 tests passing] [Test coverage] [License: MIT]

🌐 English · Русский · Español · Français · 中文 · Glossary

🟢 Just want to try the marketplace? Install nothing.

Paste https://modelmarket.dev/mcp into your MCP client and you have market_search

  • market_invoke against the live hub, with a few free trial invokes per caller and a signed receipt for each — no wallet, no key, no install. → docs/hosted-mcp-endpoint.md

Install this package when you want the other three tools (web_fetch, web_search, metis_verify), or want to point the market tools at a hub of your own.

One MCP gateway. Five hardened tools. Shared by Metis, ARGUS, and the ecosystem.

Transport: stdio (aimarket_mcp/stdio_server.py) for Glama / Claude Desktop / Cursor, built with the official Model Context Protocol Python SDK (mcp, FastMCP). Also ships Streamable-HTTP on :9090 for self-hosted deployments (aimarket-mcp-http, Docker Compose).

ItemLocation
MCP entrypoint (stdio)aimarket_mcp/stdio_server.py
MCP gateway (HTTP)aimarket_mcp/server.py
Tool handlers + securityaimarket_mcp/tools.py, aimarket_mcp/security.py
Glama / Docker (stdio)Dockerfile, glama.json
Self-host HTTPDockerfile.http, docker-compose.yml

Compatible hosts: Claude Desktop, Cursor, Glama, and any MCP client that supports stdio or Streamable-HTTP.

Tools

ToolWhat it doesHardening
web_fetchFetch a URL, return main text (readability-lite)SSRF-guarded; output sanitized + <untrusted>-wrapped
web_searchLive DuckDuckGo search → top snippetsoutput sanitized + <untrusted>
metis_verifyMetis cognition + verification envelopereturns answer + verify_score / verified gate
market_searchDiscover priced capabilities on an AIMarket hubfree; no wallet, key or channel; the hub lists only what it can execute
market_invokeRun one on the hub's free trial tierreturns the signed receipt nonce; reports the hub's 402 verbatim when the allowance is spent

Example — buy a GAIA Open-Meteo reading

text
market_search_tool(intent="gaia weather open-meteo")market_invoke_tool(  capability_id="gaia.weather.read@v1",  product_id="gaia.gateway",  source_hub="https://iot.modelmarket.dev",  # input JSON with device_id om-wx-01 — see tool schema / hub docs)

Or the same HTTP the tool wraps:

bash
curl -s -X POST "${AIMARKET_HUB_URL:-https://modelmarket.dev}/ai-market/v2/invoke" \  -H 'Content-Type: application/json' \  -H 'X-AIMarket-Sandbox-Visitor: vis_mcp_om_wx' \  -d '{"capability_id":"gaia.weather.read@v1","product_id":"gaia.gateway",       "source_hub":"https://iot.modelmarket.dev","input":{"device_id":"om-wx-01"}}'

That is the easy path (trial/sandbox). Proving an external wallet paid on-chain is a separate escrow flow (openChannel → DebitAuthorization → hub debit/settle) — see docs/onchain-journal.md §3l–§3m and gaia/docs/LIVE-RELAYS.md.

Why a gateway (not per-agent tools): generic capabilities are written once; the security core lives in one audited place. Ecosystem-specific capabilities live in their own MCP servers (aimarket-oracle-gateway, aimarket-plugins).

Configure (env)

varmeaning
AIMARKET_METIS_URLMetis verify API base (default https://metis.modelmarket.dev)
AIMARKET_METIS_KEYoptional bearer for Metis verify
AIMARKET_SEARCH_URLDuckDuckGo HTML endpoint override
AIMARKET_HUB_URLAIMarket hub for market_search / market_invoke (default https://modelmarket.dev)
AIMARKET_SANDBOX_VISITORTrial identity for market_invoke; random per install, set it to keep an allowance across reinstalls
AIMARKET_MCP_KEYHTTP only — bearer auth key
AIMARKET_MCP_PRODUCTIONHTTP only — 1 requires AIMARKET_MCP_KEY (fail-closed)
AIMARKET_MCP_RATEHTTP only — requests/min per key/IP (default 120)
AIMARKET_MCP_PORTHTTP only — listen port (default 9090)

Run (stdio — Glama / Claude Desktop)

Claude Desktop (mcpServers entry) — no clone, no working directory to get wrong:

json
{  "mcpServers": {    "aimarket-mcp": {      "command": "uvx",      "args": ["aimarket-mcp"]    }  }}

uvx fetches the published package and runs its aimarket-mcp console script, which is the stdio server. With the package already installed, "command": "aimarket-mcp" and no args does the same thing.

From a checkout, for development:

bash
pip install -e .python -m aimarket_mcp.stdio_server

Run (HTTP — self-host)

bash
pip install -e .AIMARKET_MCP_KEY=sk-... AIMARKET_MCP_PRODUCTION=1 aimarket-mcp-http   # :9090# or: docker compose up -d   (uses Dockerfile.http)

Cursor / Claude (Streamable HTTP)

json
{  "mcpServers": {    "aimarket-web": {      "type": "streamable-http",      "url": "http://127.0.0.1:9090/mcp",      "headers": {        "Authorization": "Bearer YOUR_AIMARKET_MCP_KEY"      }    }  }}

Publish on Glama

Listing: glama.ai/mcp/servers/alexar76/aimarket-mcp

Same pattern as aimarket-oracle-gateway (working on Glama): repo-root glama.json + Dockerfile + python -m aimarket_mcp.stdio_server.

FieldValue
DockerfileDockerfile (from repo — not Glama debian/uv template)
Commandpython -m aimarket_mcp.stdio_server
Placeholder parameters{}
Pinned SHAempty (latest)

Do not use the auto-generated debian:trixie-slim + uv sync + mcp-proxy -- aimarket-mcp template — that was the broken HTTP/ENOENT path.

Consumers

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

Test

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

Glama

Glama ignores repo Dockerfiles — set Build steps in admin/dockerfile:

json
["bash scripts/glama_install.sh"]

CMD: [".venv/bin/python", "-m", "aimarket_mcp.stdio_server"]. Pin main or tag glama-build (not a fixed SHA). Details: docs/GLAMA.md.

Registries

RegistryListing
Glamaglama.ai/mcp/servers/alexar76/aimarket-mcp
Official MCP Registryio.github.alexar76/aimarket-mcp — server.json + GitHub Actions
PyPIpip install aimarket-mcp
GitHub Releasesgithub.com/alexar76/aimarket-mcp/releases

來源：README.md，提交 745a82f

工具

0
工具後設資料尚未被收錄。

版本歷史

1
  1. v0.3.0最新Sep 16, 2026