---
title: CrewAI Integration
description: Use PostGrad as a tool in your CrewAI agents and crews.
---

import { Callout } from 'fumadocs-ui/components/callout';

## Overview

PostGrad integrates with CrewAI as a custom tool, giving your agents access to structured business knowledge for research, strategy, and deal evaluation tasks.

## Using the PostGrad SDK (Coming Soon)

Once the Python SDK is released, you can create a CrewAI tool:

```python
from crewai.tools import tool
from postgrad import PostGrad

client = PostGrad(api_key="pg_live_YOUR_API_KEY")

@tool("Search Business Knowledge")
def search_knowledge(query: str) -> str:
    """Search PostGrad for structured business knowledge extracted from real meetings."""
    results = client.search(query, limit=5)
    return "\n\n".join(
        f"**{e.title}** (confidence: {e.confidence})\n{e.content}"
        for e in results.data
    )
```

### Complete Crew Example

```python
from crewai import Agent, Task, Crew

researcher = Agent(
    role="Business Strategy Researcher",
    goal="Find actionable business insights from real operator experience",
    backstory="You are a senior analyst with access to a curated knowledge base of business strategy insights.",
    tools=[search_knowledge],
)

task = Task(
    description="Research pricing strategies for AI consulting services. Include confidence scores for each insight.",
    expected_output="A report with 3-5 pricing strategies ranked by confidence.",
    agent=researcher,
)

crew = Crew(agents=[researcher], tasks=[task])
result = crew.kickoff()
print(result)
```

## Using the REST API Directly

<Callout type="info">
This approach works today without waiting for the SDK.
</Callout>

```python
import requests
from crewai.tools import tool

API_KEY = "pg_live_YOUR_API_KEY"

@tool("Search Business Knowledge")
def search_knowledge(query: str) -> str:
    """Search PostGrad for structured business knowledge extracted from real meetings."""
    response = requests.get(
        "https://postgrad.io/api/v1/knowledge/search",
        headers={
            "Authorization": f"Bearer {API_KEY}",
            # "all" fans out across every subscribed feed and merges results
            # by relevance. Best default for a CrewAI tool because the agent
            # doesn't know which feed to pick.
            "X-PostGrad-Feed": "all",
        },
        params={"q": query, "limit": 5},
    )
    response.raise_for_status()
    data = response.json()

    return "\n\n".join(
        f"**{e['title']}** (from {e.get('feed_name', 'unknown feed')}, confidence: {e['confidence']})\n{e['content']}"
        for e in data["data"]
    )
```

### Multi-Agent Crew

```python
from crewai import Agent, Task, Crew

researcher = Agent(
    role="Knowledge Researcher",
    goal="Gather relevant business knowledge from PostGrad",
    backstory="You research structured business insights from real operator experience.",
    tools=[search_knowledge],
)

strategist = Agent(
    role="Strategy Advisor",
    goal="Synthesize research into actionable recommendations",
    backstory="You turn raw research into clear strategic recommendations.",
)

research_task = Task(
    description="Research what successful AI consulting firms charge and how they structure deals.",
    expected_output="Raw findings with confidence scores from the knowledge base.",
    agent=researcher,
)

strategy_task = Task(
    description="Based on the research, recommend a pricing strategy for a new AI consulting practice.",
    expected_output="A pricing strategy document with tiers, rationale, and risk factors.",
    agent=strategist,
)

crew = Crew(agents=[researcher, strategist], tasks=[research_task, strategy_task])
result = crew.kickoff()
print(result)
```