Hermes Site-Audit MCP Server
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
Hermes Site-Audit MCP Server
A self-hosted MCP server (Model Context Protocol, spec 2025-11-25+, Streamable HTTP) that gives an AI assistant honest, measurement-only website auditing superpowers — built for the Alexa+ track of the Amazon Developer Hackathon (Build, Ship, Shape, 2026).
MCP is the open standard that powers Alexa+ integrations. This server exposes four tools any MCP client (Claude Desktop, MCP Inspector, Alexa+ via MCP) can call:
Tool | What it does |
| Audits a domain for fixable technical issues — measured live: HTTP status, latency, |
| Finds public contact emails published on the site (homepage, then contact/about pages; |
| TLS certificate days-until-expiry for a domain. |
| Drafts a short outreach email quoting only measured findings, with the raw audit attached so every claim is verifiable. |
Why this exists
Freelancers and small agencies waste hours manually checking prospect websites. An assistant connected to this server can answer "audit example.com and draft me an honest outreach note" in one turn — with every claim backed by a live measurement, never invented.
Related MCP server: lilAgents MCP
Quickstart
pip install -r requirements.txt
python server.py # Streamable HTTP on http://127.0.0.1:8000/mcp
python server.py --port 9000 # custom portConnect your MCP client to http://127.0.0.1:8000/mcp using the
Streamable HTTP transport, then ask it to audit a website.
Example (via MCP Inspector or any client):
> Use audit_website on example.com
< {"domain": "example.com", "issues": ["no_meta_description"],
"issue_labels": ["homepage has no meta description (hurts click-through)"],
"score": 2, "notes": {"status": 200, "latency_s": 0.33, ...}}Tests
pip install -r requirements.txt
python test_mcp_live.pySpins up the real server on a local port and exercises all four tools through a real MCP client session (needs internet access for the live audit targets).
Project layout
server.py— MCP server (4 tools, Streamable HTTP)siteaudit_core.py— measurement-only audit engine (no MCP dependency)test_mcp_live.py— end-to-end test over real Streamable HTTP
Built with
MCP Python SDK (
mcp>=2.2, Streamable HTTP transport, spec 2025-11-25+)requestsfor the audit probes
License
MIT — see LICENSE.
Related MCP Connectors
Website health, SEO, security-header, SSL and broken-link audits for AI agents.
Live web checks for AI agents: sitemaps, robots.txt, URL status, broken links, feeds, citations.
Live SEO intelligence, rank tracking, site audits, and backlink analysis for AI agents.
Website checker for AI coding agents. Measures your site in a real browser, with evidence.
Related MCP Servers
AlicenseAqualityCmaintenanceLets any AI agent audit any website for SEO, GEO/AEO, and speed problems, reporting issues and recommendations without modifying the site.4MIT- AlicenseAqualityDmaintenanceProvides tools for AI agents to audit websites, including stack detection, DNS snapshots, and security checks.87 npmMIT
- AlicenseNot gradedqualityFmaintenanceProvides real-time website audits, lead scoring, tech-stack detection, and local-business search for AI agents doing sales outreach and competitor research.MIT
- AlicenseAqualityBmaintenanceProvides AI-visibility scoring and site auditing capabilities for websites, enabling agents to check how sites appear in AI engines like ChatGPT and Perplexity, run full SEO/security audits, and monitor changes over time.20333 npm1MIT