Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.fact0.io/llms.txt

Use this file to discover all available pages before exploring further.

​
Agno

Native Agno helpers are on the roadmap. Use the pattern below today.

​
Install

pip install fact0 agno

​
Pattern

Wrap agent.run() in an execution and emit one span per tool call. 
import fact0
from agno.agent import Agent
from agno.models.openai import OpenAIChat

client = fact0.Client(api_key="alk_live_...")

agent = Agent(
    model=OpenAIChat(id="gpt-4o"),
    tools=[...],
    markdown=True,
)

with client.telemetry.execution(agent_id="agno.research") as ex:
    with ex.span("agent.run", span_type="MODEL_INVOCATION") as span:
        response = agent.run("Summarize the Q3 earnings call")
        span.complete(output={"content": response.content[:1000]})

​
Wrap a tool

def make_audited(tool_fn):
    def wrapped(*args, **kwargs):
        with client.telemetry.current_execution().span(
            tool_fn.__name__, span_type="TOOL_CALL"
        ) as span:
            result = tool_fn(*args, **kwargs)
            span.complete(output={"result": str(result)[:500]})
            return result
    return wrapped