squirrelscan

squirrelscan is a website audit tool built for AI agents. It runs 240+ rules covering SEO, performance, security, accessibility, and more — all from a single CLI command. You can audit any site from your terminal, inside a coding agent like Claude Code or Cursor, in the cloud, or over the Model Context Protocol (MCP). The tool produces a health score, category breakdown, and fix-it guidance for each issue, outputting results in seven formats: console, HTML, markdown, LLM-optimized, JSON, XML, and plain text.

The core CLI is free and runs locally with no dependencies. Logging in unlocks cloud features that cost credits: browser rendering for JavaScript-heavy sites, AI page analysis with LLM parsability scoring and E-E-A-T authority signals, technology detection using 385 fingerprints, dead-link checking with a shared global cache, and an editor's summary written by Claude. Reports can be published to a dashboard with score history, and audits can be scheduled daily, weekly, or monthly. Every account gets 500 free credits per month, and Pro accounts cost $19/month for 3,000 credits, team support, and faster concurrent renders.

squirrelscan is best for web developers, SEO specialists, and anyone building AI-powered workflows that need structured site data. It fits naturally into CI/CD pipelines — it returns exit codes that can fail builds based on score or severity thresholds. The tool is straightforward: install with one curl command, run an audit, and get actionable results. It doesn't try to be a full SEO platform or analytics suite — it's a focused audit engine that speaks the language of both humans and agents.