CrewAI Integration

Combine Agent Control security with CrewAI Guardrails for production customer support.

Combines Agent Control (security and compliance) with CrewAI Guardrails (quality retries) for production customer support.

What It Does

• Agent Control (security): PRE, POST, and FINAL blocks unauthorized access and PII.
• CrewAI Guardrails (quality): validates length, structure, and tone with up to three retries.

Prerequisites

Before running this example, ensure you have:

• Python 3.12+
• uv (fast Python package manager)
• Docker (for PostgreSQL required by Agent Control server)

Installation

1. Install Monorepo Dependencies

From the monorepo root, install all workspace packages:

cd /path/to/agent-control
make sync

This installs the Agent Control SDK and all workspace packages in editable mode.

2. Install CrewAI Example Dependencies

Navigate to the CrewAI example and install its specific dependencies:

cd examples/crewai
uv pip install -e . --upgrade

3. Set OpenAI API Key

Create a .env file or export the environment variable:

export OPENAI_API_KEY="your-key-here"

4. Start the Agent Control Server

In a separate terminal, start the server from the monorepo root:

cd /path/to/agent-control
make server-run

Verify the server is running:

curl http://localhost:8000/health

5. Setup Content Controls (One-Time)

From the examples/crewai directory, run the setup script:

uv run python setup_content_controls.py

Running the Example

Make sure you’re in the examples/crewai directory and run:

uv run python content_agent_protection.py

Expected Behavior

Scenario	Layer	Result
Unauthorized access	Agent Control PRE	Blocked
PII in tool output	Agent Control POST	Blocked
Short or low-quality response	Guardrails	Retry then pass
Agent bypass attempt	Agent Control FINAL	Blocked

Output Legend

• PRE checks input before the LLM
• POST checks tool output for PII
• FINAL checks the crew’s final response
• Agent Control blocks immediately (no retries), violations are logged
• Guardrails retry with feedback (quality only)

Agent Control and CrewAI Integration

Agent Control works with CrewAI’s orchestration:

1. CrewAI agent layer: plans tasks, selects tools, manages conversation flow
2. Agent Control layer: enforces controls and business rules at tool boundaries

User Request
    ↓
CrewAI Agent (planning & orchestration)
    ↓
Decides to call tool
    ↓
@control() decorator (PRE-execution)  ← LAYER 1: Validates input
    ↓
Tool executes (LLM generation)
    ↓
@control() decorator (POST-execution) ← LAYER 2: Validates tool output
    ↓
If blocked, agent may generate own response
    ↓
Final Output Validation              ← LAYER 3: Validates crew output (bypass protection)
    ↓
Return to user (or block if control violated)

Source Code

View the complete example with all scripts and setup instructions:

CrewAI Integration Example