Skip to main content
Fallom integrates seamlessly with Agno to trace all your agent LLM calls automatically.
Get your API key from the dashboard.

Installation

npm install @fallom/trace agno

Quick Start

import fallom from "@fallom/trace";
import { Agent } from "agno";

// Initialize Fallom
await fallom.init({ apiKey: "your-api-key" });

// Set session context
fallom.trace.setSession("my-agent", sessionId);

// Create your Agno agent
const agent = new Agent({
  model: "gpt-4o",
  instructions: "You are a helpful assistant."
});

// All LLM calls within the agent are traced
const response = await agent.run("What is the capital of France?");

Model A/B Testing with Agno

Test different models with your Agno agents:
import fallom, { models } from "@fallom/trace";
import { Agent } from "agno";

await fallom.init({ apiKey: "your-api-key" });

// Get assigned model for this session
const modelId = await models.get("agno-agent", sessionId, {
  fallback: "gpt-4o"
});

fallom.trace.setSession("agno-agent", sessionId);

// Use the assigned model with your agent
const agent = new Agent({
  model: modelId,
  instructions: "You are a helpful assistant."
});

const response = await agent.run("Explain quantum computing");

Prompt Management with Agno

Use managed prompts for your agent instructions:
import fallom, { prompts } from "@fallom/trace";
import { Agent } from "agno";

await fallom.init({ apiKey: "your-api-key" });

// Get managed prompt for agent instructions
const prompt = await prompts.get("agent-instructions", {
  variables: { persona: "helpful assistant", domain: "customer support" }
});

fallom.trace.setSession("agno-agent", sessionId);

const agent = new Agent({
  model: "gpt-4o",
  instructions: prompt.system
});

const response = await agent.run("How do I reset my password?");

Next Steps