AgentReady | AI Agents Can't Read Your Website — Fix It Free

# AgentReady > AgentReady is an agent interface platform that makes any website AI-agent native. Paste any public URL and AgentReady generates a spec-compliant llms.txt, a hosted MCP server, cited RAG answers, and an agent-readiness grade — no code or developer required. It is live at https://agentready.it.com. ## What AgentReady Does AgentReady solves the problem that AI agents (ChatGPT plugins, Claude, Cursor, Perplexity) cannot reliably query most websites: they parse raw HTML, miss JavaScript-rendered content, hallucinate answers, and have no way to know what a site supports. AgentReady provides a structured interface layer between agents and websites. Key facts about AgentReady: - Founded 2026; headquarters: remote - Free plan available; Pro at $49/month; Team at $149/month - MCP endpoint (Streamable HTTP): https://agentready.it.com/api/mcp - Site-specific ask endpoint: POST https://agentready.it.com/api/sites/{siteId}/ask with body {"query": "..."} - Site-specific MCP: https://agentready.it.com/api/sites/{siteId}/mcp - llms.txt for any indexed site: https://agentready.it.com/api/sites/{siteId}/llms.txt - npm CLI package: @agentreadyweb/mcp (no install: `npx @agentreadyweb/mcp`) - AI model used for answer generation: Anthropic claude-sonnet-4-6 - Embeddings: OpenAI text-embedding-3-small (pgvector on Supabase) - 152 sites currently indexed in the public directory ## Pricing - **Free** ($0/month): 1 site, 100 pages crawled, 5 index refreshes/month, public MCP endpoint, llms.txt generation, cited RAG answers, agent-readiness grade, 7-day analytics, query intent analytics, structured fact extraction (read-only), crawl changelog (last 3 entries), answer signing - **Pro** ($49/month): 5 sites, 500 pages per site, 50 refreshes/month, everything in Free, 30-day analytics, private MCP with API key auth, GitHub check runs on every refresh, Slack regression alerts, Agent Simulator (5-task scoring), query gap detection with AI-drafted content briefs, answer tests

{"openapi":"3.1.0","info":{"title":"AgentReady API","version":"1.0.0","description":"HTTP interfaces for discovering indexed sites, querying documentation, grading agent-readiness, and submitting crawls. Ask responses include retrieval metadata and may be served from a short-lived cache. Responses include X-Request-Id for tracing; metered endpoints include RateLimit-Limit, RateLimit-Remaining, RateLimit-Reset, and Retry-After when limited."},"servers":[{"url":"https://agentready.it.com"}],"paths