Support

How can we help?

Pick a channel below, or check the FAQ — most scanner questions are answered there or in the documentation.

Frequently asked questions

What is the Agent Readiness Score (ARS)?

ARS 1.0 is a versioned, weighted composite score (0–100) measuring how well AI agents can discover, understand, and operate your product. It spans eight dimensions: Discoverability, Machine Entrypoints, Canonical Sources, Content Quality, Graph Connectivity, Journey Success, Freshness, and Runtime Validity. Bands: 85+ AI-Native, 70–84 AI-Ready, 50–69 AI-Capable, below 50 AI-Limited.

Is the scanner free?

Yes. Running a scan at scan.glintbase.dev is free, and the scanner, CLI, MCP server, and agent skill are open source at github.com/glintbase/glintscanner.

How do I improve my score?

Every scan report includes prioritized remediation advice with estimated effort and score impact. Common wins: publish llms.txt and a real docs root, add code examples to key pages, fix dead links, and expose a changelog and status surface. The get_remediation MCP tool produces the same guidance programmatically.

Can I run scans in CI/CD?

Yes — use npx @glintbase/cli scan <url> --fail-under <score> to gate deploys on a minimum Agent Readiness Score. See /docs for full CLI usage.

Does scanning require API keys?

No. The hosted scanner, the CLI's core pipeline, and all nine MCP tools run without API keys.

How do I remove or re-run a scan of my site?

Re-run a scan any time at scan.glintbase.dev — the leaderboard reflects the latest result. For removal requests or anything else, email contact@glintbase.dev.