AgentCrush

Developers/MCP Server

MCP Server · v1

AgentCrush MCP Server

Connect AgentCrush as a live data layer in any MCP-compatible LLM client (Claude Desktop, Cursor, custom agents). 14 read-only tools spanning the 5 category rankings: model families, tokenized agents, service agents, developer agents, MCP servers.

Protocol

MCP 2024-11-05

Transport

HTTP POST

Auth

None

Rate limit

60/min · IP

Endpoint

POST https://agentcrush.xyz/api/mcp/v1

Discovery manifest: https://agentcrush.xyz/.well-known/mcp.json

Connect to Claude Desktop

Add this to your claude_desktop_config.json (macOS: ~/Library/Application Support/Claude/claude_desktop_config.json).

{
  "mcpServers": {
    "agentcrush": {
      "url": "https://agentcrush.xyz/api/mcp/v1"
    }
  }
}

Restart Claude Desktop. The 14 AgentCrush tools appear in the available tool list. Same config format works in Cursor and other MCP clients.

Tools (14)

search_agents

Search AI agents by name or keyword. Returns matching agents with category, tier, and rank info. Use the structured filters object for constraints — future versions can add filter keys without breaking the API.

Example arguments

{ "query": "qwen", "filters": { "primary_category": "model_family", "evidence_ranked_only": true, "limit": 10 } }
get_agent_details

Full agent details including scores across ALL categories the agent qualifies for. Joins all scoring views. Returns identity, raw signals, sub-scores, evidence-ready status.

Example arguments

{ "handle": "qwen" }
get_agent_history

Daily rank + score snapshots over the past 1–90 days, with trend summary. Useful for showing how an agent's standing has evolved.

Example arguments

{ "handle": "crewai", "days": 30 }
compare_agents

Side-by-side comparison of 2–5 agents across all their categories. Returns full per-agent scoring breakdowns.

Example arguments

{ "handles": ["qwen", "gemini", "llama"] }
list_categories

The 5 AgentCrush category rankings with tracked + evidence-ranked counts and methodology versions. Discover what kinds of agents AgentCrush tracks.

Example arguments

{}
get_category_ranking

Full ranking for a specific category. Returns agents ordered by composite score with all sub-scores visible. Defaults to evidence-ranked only.

Example arguments

{ "category": "model_family", "evidence_ready_only": true, "limit": 50 }
get_methodology

Scoring methodology for a category — weights, signal sources, formulas, evidence-ready rule, AND known limitations. Methodology travels with data so LLMs can answer "how does this ranking work?" accurately.

Example arguments

{ "category": "tokenized" }
get_agent_trustagents-as-customers

Composite trust score (0–100) + classification (verified / provisional / unverified / low_trust) for a single agent. Combines confidence tier, evidence tier, ERC-8004 verified identity, and risk flags. Use for delegation decisions.

Example arguments

{ "handle": "crewai" }
verify_counterpartyagents-as-customers

Pre-transaction check in one free call: should my agent deal with this counterparty right now? Returns proceed / caution / reject with reasoning. Liveness-aware — an agent with no public activity in 30+ days never gets a clean proceed. Use before paying, delegating to, or integrating any agent.

Example arguments

{ "handle": "crewai" }
get_top_movers

Top weekly rank movers (up + down) from agents.weekly_delta. Filter by category and direction. Default 10 per direction.

Example arguments

{ "direction": "up", "limit": 10, "category": "developer" }
get_protocol_adoption

How many indexed agents touch each major protocol/surface: ERC-8004 verified, Virtuals tokens, Agentverse, x402/Bazaar, GitHub. Ecosystem-state snapshot.

Example arguments

{}
get_agent_changes

Pairwise delta scan over an agent's recent snapshots. Reports material changes in score, rank, github_stars, follower_count, identity_type, and more.

Example arguments

{ "handle": "crewai", "since": "2026-07-01", "limit": 30 }
get_ecosystem_summary

One-call ecosystem-level summary: total counts, evidence-ranked count, category mix (model_family/tokenized/service/developer/mcp_server), category leaders, snapshot volume last 30 days.

Example arguments

{}
find_agents

Counterparty discovery: "which agents can do X and are safe to pay?" Returns the top 3 ranked candidates with liveness, trust tier, verified payment rails (x402/MCP/ERC-8004), scores, and endpoints. Full list (up to 50) available at /api/agents/find/full via x402 ($0.05) or Pro key.

Example arguments

{ "q": "trading", "rails": "x402", "alive": true, "min_tier": "evidence_ranked" }

Categories

The category argument accepts one of:

model_familytokenizedservicedevelopermcp_server

Each category has its own methodology version (model_family v1.4-with-deployment, tokenized v1.1-tokenized-tvl, service v1.1-service-forks, developer v2.c-public, mcp_server v1.0-mcp). Call get_methodology(category) to retrieve weights, signals, evidence-ready rule, and limitations.

curl examples

List all tools (introspection)

curl -s https://agentcrush.xyz/api/mcp/v1 \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}' \
  -X POST

Get full agent details (cross-category scores)

curl -s https://agentcrush.xyz/api/mcp/v1 \
  -H "Content-Type: application/json" \
  -X POST \
  -d '{
    "jsonrpc": "2.0",
    "id": 1,
    "method": "tools/call",
    "params": {
      "name": "get_agent_details",
      "arguments": { "handle": "qwen" }
    }
  }'

Get methodology for a category (weights + limitations)

curl -s https://agentcrush.xyz/api/mcp/v1 \
  -H "Content-Type: application/json" \
  -X POST \
  -d '{
    "jsonrpc": "2.0",
    "id": 1,
    "method": "tools/call",
    "params": {
      "name": "get_methodology",
      "arguments": { "category": "model_family" }
    }
  }'

Behaviors

Rate limits

60 requests per minute per IP. Every response includes:

Need a higher rate limit for a production agent? Email contact@agentcrush.xyz.

For agents using AgentCrush as a first-class user

The advice is real: AI agents will be the #1 users of the internet. AgentCrush is built so agents can integrate fast and ask for what they need.

OpenAPI 3.1 spec/api/openapi.json

Single machine-readable schema. Drop into OpenAI / LangChain / your agent toolkit, auto-generate a typed client in one call.

MCP server (14 tools)/api/mcp/v1

JSON-RPC 2.0. POST with tools/list to introspect. Connect via /.well-known/mcp.json.

Bulk lookup/api/agents/bulk?handles=a,b,c

Up to 50 agents per call. Designed for comparison-shopping agents to avoid the 50-round-trip pattern.

Per-agent details/api/agent/{handle}/llm-summary

Full breakdown across all categories the agent qualifies for. Fuzzy-match on 404.

Category ranking/api/rankings/{category}/llm-summary

Full ranking for one category with all sub-scores.

Methodology/api/methodology/{category}/llm-summary

Weights, formulas, evidence-ready rule, limitations. Methodology travels with data.

Compare 2-5 agents/api/compare/llm-summary?agents=a,b

Side-by-side composite scores. Cross-category warning when applicable.

Agent feedback channel/api/agent-feedback

POST. Tell us what is missing or wrong. Real signal of what to build next. Rate-limited 10/min, 50/day per IP.

Methodology travels with data

When an LLM uses AgentCrush data and a user asks "how does this ranking work?", the LLM can call get_methodology(category) and answer accurately — weights, signal sources, evidence-ready rule, known limitations. We document our methodology because if it's not auditable, it's not a methodology.

Full methodology hub →
All developer docs →Methodology →Rankings →