AgentForge Trust MCP gives any AI agent a single MCP connection that exposes a catalog of 100+ MCP servers — trust-scored, searchable, and executable. Trust Scores (0-100) are computed across five dimensions: code health, security scan, behavioral audit, community trust, and EU compliance. Always call check_trust before connecting to a new MCP server. v0.3.0 ships 8 tools (4 trust + 4 action) plus a P0 RCE sanitization layer (input validation with prototype pollution defense, call guards with rate limiting and timeout, runtime anomaly detection for SQL/CMD/SSRF probes). Zero external runtime dependencies. 27/27 tests passing. Install: npx -y agentforge-trust-mcp@latest. Trust tools (check_trust, evaluate_policy, list_trusted, recommend) require no auth. Action tools (broadcast_search, call_tool) auto-issue API key via register_agent — no signup form.
This server has not been audited yet. Trust score will appear once the first audit completes.
Search across multiple MCP servers in one call (requires AGENTFORGE_API_KEY).
{
"type": "object",
"required": [
"query"
],
"properties": {
"query": {
"type": "string"
},
"max_servers": {
"type": "integer",
"default": 5
}
}
}Invoke any tool on any server with proxied execution, rate limiting, and billing (requires AGENTFORGE_API_KEY).
{
"type": "object",
"required": [
"server",
"tool"
],
"properties": {
"tool": {
"type": "string"
},
"input": {
"type": "object"
},
"server": {
"type": "string"
}
}
}Fetch the AgentForge Trust Score for an MCP server identified by server_id, slug, or github_url. Always call before connecting to a new server.
{
"type": "object",
"properties": {
"slug": {
"type": "string"
},
"server_id": {
"type": "string"
},
"github_url": {
"type": "string"
}
}
}Browse the AgentForge catalog with full-text or semantic search.
{
"type": "object",
"properties": {
"q": {
"type": "string"
},
"limit": {
"type": "integer",
"default": 20
},
"category": {
"type": "string"
},
"semantic": {
"type": "boolean"
}
}
}AgentForge Trust MCP gives any AI agent a single MCP connection that exposes a catalog of 100+ MCP servers — trust-scored, searchable, and executable. Trust Scores (0-100) are computed across five dimensions: code health, security scan, behavioral audit, community trust, and EU compliance. Always call check_trust before connecting to a new MCP server. v0.3.0 ships 8 tools (4 trust + 4 action) plus a P0 RCE sanitization layer (input validation with prototype pollution defense, call guards with rate limiting and timeout, runtime anomaly detection for SQL/CMD/SSRF probes). Zero external runtime dependencies. 27/27 tests passing. Install: npx -y agentforge-trust-mcp@latest. Trust tools (check_trust, evaluate_policy, list_trusted, recommend) require no auth. Action tools (broadcast_search, call_tool) auto-issue API key via register_agent — no signup form.
AgentForge Trust provides 8 tools including broadcast_search, call_tool, check_trust, and 5 more.
Yes, AgentForge Trust is completely free to use with no usage limits on the free tier.
Allow/deny gate using a policy (min scores, required/forbidden badges).
{
"type": "object",
"required": [
"policy"
],
"properties": {
"slug": {
"type": "string"
},
"policy": {
"type": "object"
},
"server_id": {
"type": "string"
}
}
}Filtered list of servers matching a policy.
{
"type": "object",
"properties": {
"limit": {
"type": "integer",
"default": 50
},
"category": {
"type": "string"
},
"min_overall": {
"type": "number"
}
}
}Search servers by use case with trust filtering.
{
"type": "object",
"required": [
"query"
],
"properties": {
"limit": {
"type": "integer",
"default": 10
},
"query": {
"type": "string"
},
"min_overall": {
"type": "number"
}
}
}Self-register this agent with AgentForge. Returns an af_agent_* API key. No signup form, idempotent on slug.
{
"type": "object",
"required": [
"name"
],
"properties": {
"name": {
"type": "string"
},
"description": {
"type": "string"
},
"capabilities": {
"type": "array"
}
}
}AgentForge Trust is listed under the developer-tools category in the AgentForge MCP registry.
To connect AgentForge Trust, click the "Connect Agent" button on this page to get the configuration snippet. Add it to your MCP client (Claude Desktop, Cursor, or any MCP-compatible tool). Your AI agent will then have access to all of AgentForge Trust's tools via the Model Context Protocol.