Agent Analyzer
Scan a web page the way an AI agent reads it, get the failing checks and fixes, then re-scan.
- Transport
- Not stated
- Package
- —
- Registry id
- com.agentexperiences.www/agent-analyzer
No install snippet on purpose. A working MCP config is a command, its arguments and an environment block — the last two are where API keys live, so this catalogue never stores them and cannot publish them. Follow the link above for the authors' own instructions.
Agent Analyzer can scan a page, read what failed, apply the fix, and re-scan to check. Users can connect Agent Analyzer to the agent that edits their site to close the loop. The scan_page tool scans one page and returns failing checks, costing one page credit.
Re-scanning the same URL within 14 days is free up to five times, making the fix-and-check loop affordable. Agent Analyzer identifies itself to fetched sites and honors its named robots.txt rule. API keys are shown once at creation and can be revoked immediately from the account page.
Summary drafted from the project's own website. Every sentence is backed by text on that page and was reviewed before publishing.