AgentCrush
Look up rankings, methodology, and side-by-side comparisons before you commit to an agent framework or model stack.
Overview
AgentCrush is an MCP server for the Idea phase that supplies rankings, search, and comparisons across the AI agent economy via 7 tools.
What is this MCP server?
- 7 MCP tools for agent rankings, search, compare, methodology, and category listing
- Documented flows: get_agent_details, get_category_ranking, search_agents, compare_agents, get_methodology
- Hosted streamable remote at https://www.agentcrush.xyz/api/mcp/v1
- OpenAPI spec and methodology URL for citable ranking explanations
- Rate limit 60 requests per minute per IP
- 7 MCP tools documented for rankings, search, compare, and methodology
- OpenAPI at https://www.agentcrush.xyz/api/openapi.json
What problem does it solve?
Builders waste days guessing which agent frameworks are credible because market lists are scattered and methodology is opaque.
Who is it for?
Solo builders evaluating agent stacks, writing positioning, or answering client questions with comparable, sourced rankings.
Skip if: Running production agents, executing code on devices, or teams that already standardized on one stack with no evaluation need.
What do I get? / Deliverables
Your agent pulls ranked, methodology-backed agent intelligence so you scope and build on evidence instead of hype.
- Agent detail and category ranking answers grounded in AgentCrush methodology
- Side-by-side agent comparisons suitable for PRD or architecture notes
Recommended MCP Servers
Journey fit
Choosing among CrewAI, LangGraph, and similar stacks is research work that happens before you validate and build. AgentCrush is research intelligence: category rankings, search, compare_agents, and methodology tools for the agent economy.
How it compares
Agent-economy intelligence MCP with 7 research tools, not an implementation framework or deployment server.
Common Questions / FAQ
Who is AgentCrush for?
Developers and indie founders researching AI agents who want MCP-accessible rankings, search, and comparisons from AgentCrush.
When should I use AgentCrush?
When a user or your agent needs agent details, category rankings, comparisons, or methodology before choosing tooling.
How do I add AgentCrush to my agent?
Connect the remote MCP client to https://www.agentcrush.xyz/api/mcp/v1 using the discovery manifest at /.well-known/mcp.json and docs at /developers/mcp.