Examples →

Fetch a Stack Overflow question for an AI agent

Stack Overflow's HTML is well-structured, so AgentFetch's local Trafilatura pass extracts the question, all answers, and code blocks cleanly without hitting their rate limits.

Try it

Get a free key at agentfetch.dev (500 free fetches/mo, no credit card).

curl -X POST https://api.agentfetch.dev/fetch \
  -H "Content-Type: application/json" \
  -H "X-AgentFetch-Key: af_xxx" \
  -d '{
    "url": "https://stackoverflow.com/questions/12345/how-to-x",
    "max_tokens": 2000
  }'

In an MCP-aware agent

fetch_url(
  url="https://stackoverflow.com/questions/12345/how-to-x",
  max_tokens=2000,
)

In LangChain

from langchain_agentfetch import AgentFetchTool

tool = AgentFetchTool(api_key="af_xxx")
result = tool.run({"url": "https://stackoverflow.com/questions/12345/how-to-x"})

How it routes

AgentFetch automatically routes fetch stack overflow-style URLs through Trafilatura based on domain heuristics — you don't pick a fetcher.

When agents reach for this

Coding agents, debugging assistants, RAG over SO knowledge.

Related patterns

Fetch a Hacker News thread for an AI agent
Fetch a Twitter / X thread for an AI agent
Fetch a GitHub README for an AI agent
Fetch an arXiv paper for an AI agent
Fetch and extract text from a PDF for an AI agent
Fetch a public Notion page for an AI agent