# Pickable > Pickable is a public website analysis demo. It combines agent-readiness observations from Ora and Cloudflare's isitagentready.com with basic marketing visibility checks. Scores describe available evidence; they do not guarantee task completion, referrals, rankings, or revenue. ## Product - [Pickable home](https://solveny.dev/): enter a public domain to request an asynchronous scan. - [About Pickable](https://solveny.dev/about): purpose, sources, and score limitations. - [Demo and limits](https://solveny.dev/pricing): free public demo and configured resource limits. ## Documentation - [Getting started and interpretation](https://solveny.dev/docs) - [REST and MCP API guide](https://solveny.dev/api) - [Agent integration guide](https://solveny.dev/docs/agent-scan) - [Authentication and CSRF notes](https://solveny.dev/auth.md) - [Full agent reference](https://solveny.dev/llms-full.txt) - [Documentation-specific index](https://solveny.dev/docs-llms.txt) - [API-specific index](https://solveny.dev/api-llms.txt) ## Machine-readable resources - [OpenAPI 3.1](https://solveny.dev/openapi.json) - [MCP server card](https://solveny.dev/.well-known/mcp/server-card.json) - [ARD manifest](https://solveny.dev/.well-known/ard.json) - [Agent discovery document](https://solveny.dev/.well-known/agent.json) - [Agent Skills index](https://solveny.dev/.well-known/agent-skills/index.json) - [Pickable scan skill](https://solveny.dev/skills/pickable-scan/SKILL.md) - [RFC 9727 API catalog](https://solveny.dev/.well-known/api-catalog) ## Important behavior - Scan creation is asynchronous. Poll the scan status endpoint and treat a report received while sources are running as partial. - Scan history, statuses, reports, previews, and intelligence reads are scoped to the caller’s direct peer IP address. Forwarded-IP headers are not trusted; clients behind a reverse proxy may share the same scope. This is not account authentication or a confidentiality guarantee. Provider evidence is redacted and raw-source endpoints return 404. - REST scan creation uses a CSRF token from GET /api/config or a configured operator-issued bearer key. - MCP read tools are available by default. MCP scan creation requires both server-operator opt-in and bearer authorization. - The MCP HTTP endpoint accepts POST requests; use the agent integration guide or server card for connection details. - Follow Retry-After when scan submission returns 429.