{"name":"agent-seo","description":"How do you know if an AI agent is any good before you use it? And if you're building one, how do you know it's discoverable? There are 56,000+ MCP servers — none let another agent evaluate them before committing. agent-seo solves both: it scores any AI agent on 5 trust dimensions (Schema Quality, Reliability, Developer Experience, Ecosystem Signal, Maintenance Health) and tells you exactly what's missing and how to fix it. Use it to evaluate agents before integrating, or to improve your own agent before releasing. Think of it as SEO for agents — the infrastructure that makes agents evaluable, not just discoverable.","url":"https://chic-empathy-production-7c2e.up.railway.app","version":"0.5.0","provider":{"name":"Manav Agarwal","url":"https://github.com/manavaga","repository":"https://github.com/manavaga/agent-seo"},"repository":"https://github.com/manavaga/agent-seo","capabilities":[{"name":"score_agent","description":"Before integrating an AI agent, check if it's trustworthy. Scores any MCP server or AI agent endpoint on 5 dimensions: How well does it describe its tools? Does the MCP handshake work? Is it documented? Do others use it? Is it maintained? Returns a 0-100 score with letter grade and a prioritized list of fixes showing exactly what to add and how many points each improvement is worth."},{"name":"compare_agents","description":"Choosing between two AI agents for a task? Scores both and shows which one is stronger in each category — schema quality, reliability, documentation, ecosystem, and maintenance. Helps you pick the more trustworthy option based on data, not guesswork."},{"name":"get_fix_recommendations","description":"Building an AI agent and want to make it more discoverable and trustworthy? Get a prioritized list of improvements: what to add, expected point gain, code templates, and links to the relevant specs. Shows your current score and what it would be after each fix."}],"protocols":{"mcp":"https://chic-empathy-production-7c2e.up.railway.app/mcp","a2a":"https://chic-empathy-production-7c2e.up.railway.app/.well-known/agent.json"},"pricing":"Free — no authentication required","update_frequency":"Real-time scoring on demand","response_format":"application/json","use_cases":["Evaluate an MCP server before integrating it into your workflow","Check your own agent's trust score during development","Compare competing agents to pick the best one for a task","Get actionable fixes to improve your agent's discoverability","Audit agent endpoints for compliance with A2A and MCP standards"]}