About agentgauge
AgentGauge scans a website’s readiness for AI search and AI agents, grading it against the AgentReady standard. Its Claude Code skill can prioritize findings, make codebase fixes, and rescan to verify them.
Key features
- 01Grade AI search discoverability
- 02Check AI agent usability
- 03Run 31 checks across six categories
- 04Prioritize findings by conformance level
- 05Apply fixes and rescan
Use cases
- 01Audit a site’s AI readiness
- 02Find gaps in AI search visibility
- 03Improve agent access to site capabilities
- 04Verify fixes after deployment
MCP support
The remote MCP server exposes a `scan_website` tool that scans a URL without authentication. It can be connected from Claude Code, Cursor, or Claude Desktop.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://mcp.agentgauge.ai/mcpclaude mcp add --transport http glasscaseofemotion-agentgauge https://mcp.agentgauge.ai/mcp{"mcpServers":{"glasscaseofemotion-agentgauge":{"type":"http","url":"https://mcp.agentgauge.ai/mcp"}}}Open in Cursor{"servers":{"glasscaseofemotion-agentgauge":{"type":"http","url":"https://mcp.agentgauge.ai/mcp"}}}Open in VS Code{"mcpServers":{"glasscaseofemotion-agentgauge":{"type":"http","url":"https://mcp.agentgauge.ai/mcp"}}}{"mcpServers":{"glasscaseofemotion-agentgauge":{"serverUrl":"https://mcp.agentgauge.ai/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
Tools
1| Parameter | Type | Required | Description |
|---|---|---|---|
| url | string | Yes | The website URL or domain to scan, e.g. "example.com" |
Details
From the project READMEAgentGauge skill for Claude Code
Scan any website's AI readiness from inside your AI coding tool — then let it fix the findings in your codebase and re-scan to verify.
AgentGauge grades whether AI search can find you (citability, llms.txt, schema, AI-crawler access) and whether AI agents can use you (tools, identity, commerce signals), against the AgentReady standard with stable AR-IDs and MUST/SHOULD/MAY conformance. Free scan, instant letter grade.
Install
1. The MCP server (recommended)
claude mcp add --transport http agentgauge https://mcp.agentgauge.ai/mcp
Cursor / Claude Desktop JSON config:
{
"mcpServers": {
"agentgauge": { "url": "https://mcp.agentgauge.ai/mcp" }
}
}
No auth. One tool: scan_website({ url }).
2. The skill
Copy SKILL.md into your Claude Code skills directory as agentgauge/SKILL.md:
mkdir -p ~/.claude/skills/agentgauge
curl -fsSL https://raw.githubusercontent.com/GlassCaseOfEmotion/agentgauge-skill/main/SKILL.md \
-o ~/.claude/skills/agentgauge/SKILL.md
Use
/agentgauge scan mysite.com
or just ask: "check my site's AI readiness and fix what's failing". The skill scans, prioritizes the findings (spec MUSTs first), implements fixes in your repo, and re-scans after deploy to verify the grade actually moved.
What gets checked
31 checks across 6 categories: Discoverability, Content for Agents, AI Visibility (GEO), Capabilities (MCP/A2A/OpenAPI), Identity & Access, and Commerce.
Built by AgentGauge · Weekly monitoring: Watch by AgentGauge · Contact: [email protected]
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.