pydantic-ai-contextfetch

ContextFetch web search and page reading tools for Pydantic AI, as a toolset.

pip install pydantic-ai-contextfetch
export CONTEXTFETCH_API_KEY="your-api-key"  # optional: without a key, requests use the keyless tier

from pydantic_ai import Agent

from pydantic_ai_contextfetch import ContextFetchToolset

agent = Agent('openai:gpt-5.2', toolsets=[ContextFetchToolset()])
result = agent.run_sync('What changed in the latest Python release?')
print(result.output)

Tools

• contextfetch_search(query, include_domains?, exclude_domains?, start_published_date?): web search. Returns a list of {title, url, published_date, content}, where content is the page as Markdown: by default only the parts relevant to the query.
• contextfetch_read(url, query?): one page as Markdown. With query, only the parts relevant to it. A bad or unreachable URL raises ModelRetry, so the model can correct it.

Options

ContextFetchToolset(
    api_key=None,              # defaults to CONTEXTFETCH_API_KEY
    num_results=5,             # results per search, 1 to 25
    content='pruned_context',  # or 'text' for whole pages
    id=None,                   # toolset id for durable execution
)

To expose only one tool, use Pydantic AI's toolset filtering: ContextFetchToolset().filtered(lambda ctx, tool: tool.name == 'contextfetch_search').

Tests

uv run --group test pytest tests                                   # unit tests, no network
CONTEXTFETCH_API_KEY=... uv run --group test pytest tests/test_live.py  # live, 2 credits

MIT licensed.