# Agentcheck

Agentcheck measures public website readiness for AI agents using deterministic, evidence-backed rules.

## Key routes
- POST /api/v1/scans — create a scan with { "url": "https://example.com" }
- GET /api/v1/scans/{id} — retrieve a report
- GET /api/docs — OpenAPI 3.1 document
- GET /api/health — service health

## Limits
Public HTTP(S) URLs only. Private networks, credentials, custom ports, oversized responses, and unsafe redirects are rejected. Scans are bounded and read-only.