AFagentfeeddocs
AFagentfeeddocs

Getting Started

QuickstartAuthentication

Integration

MCP SetupREST APIPython SDKTypeScript SDK

REST API

Query AgentFeed via HTTP endpoints.

Base URL

https://api.agentfeed.live/v1

Endpoints

POST /query

Search across expert sources.

curl -X POST https://api.agentfeed.live/v1/query \
  -H "Authorization: Bearer af_sk_..." \
  -H "Content-Type: application/json" \
  -d '{
    "query": "What do analysts think about Apple Vision Pro?",
    "categories": ["tech", "finance"],
    "freshness": "7d",
    "max_results": 5
  }'

Response:

{
  "results": [
    {
      "article_id": "uuid",
      "title": "Vision Pro and the Spatial Computing Bet",
      "source": "Stratechery",
      "author": "Ben Thompson",
      "published_at": "2026-04-12T08:00:00Z",
      "freshness_hours": 48,
      "relevance_score": 0.96,
      "snippet": "The real question for Vision Pro isn't unit sales...",
      "source_url": "https://stratechery.com/..."
    }
  ],
  "query_cost_usd": 0.004,
  "latency_ms": 43
}

GET /sources

List available expert sources.

curl https://api.agentfeed.live/v1/sources \
  -H "Authorization: Bearer af_sk_..."

GET /article/:id

Get full article content in markdown.

curl https://api.agentfeed.live/v1/article/uuid \
  -H "Authorization: Bearer af_sk_..."

MCP Setup

Connect AgentFeed to Claude Desktop, Cursor, or any MCP client.

Python SDK

Use AgentFeed from Python.

On this page

Base URLEndpointsPOST /queryGET /sourcesGET /article/:id