Skip to main content

​
What You’ll Learn

  • Defining a @tool with crewai.tools that executes Python in a Declaw sandbox
  • Creating a CrewAI Agent, Task, and Crew
  • Demo mode that exercises the sandbox tool directly without needing an OpenAI key

​
Prerequisites

  • Declaw instance running and DECLAW_API_KEY / DECLAW_DOMAIN set
  • OPENAI_API_KEY (optional — the example runs in demo mode without it)
pip install declaw python-dotenv crewai
This example is available in Python. TypeScript support coming soon.

​
Code Walkthrough

​
1. Define the Declaw sandbox tool

Use the crewai.tools @tool decorator with a display name as the first argument: 
from crewai.tools import tool
from declaw import Sandbox

@tool("Execute Python Code")
def execute_python(code: str) -> str:
    """Execute Python code securely in a Declaw sandbox and return the output."""
    sbx = Sandbox.create(template="python", timeout=300)
    try:
        sbx.files.write("/tmp/code.py", code)
        result = sbx.commands.run("python3 /tmp/code.py", timeout=30)
        return f"{result.stdout}\n{result.stderr}".strip()
    finally:
        sbx.kill()
CrewAI uses the docstring as the tool’s description when presenting it to the LLM. Keep it concise and accurate.

​
2. Create an Agent with the tool

from crewai import Agent

coder = Agent(
    role="Python Developer",
    goal="Write and execute Python code to solve tasks",
    backstory="You are an expert Python developer with access to a secure sandbox.",
    tools=[execute_python],
    verbose=True,
)

​
3. Define a Task and run the Crew

from crewai import Task, Crew

task = Task(
    description="Compute the sum of all prime numbers below 50 and print the result.",
    expected_output="The sum of all prime numbers below 50.",
    agent=coder,
)

crew = Crew(agents=[coder], tasks=[task], verbose=True)
result = crew.kickoff()
print(f"Final result: {result}")

​
4. Demo mode (no API key needed)

Run the sandbox tool directly to verify it works before hooking it up to CrewAI: 
from declaw import Sandbox

code = """\
def is_prime(n):
    if n < 2:
        return False
    for i in range(2, int(n**0.5) + 1):
        if n % i == 0:
            return False
    return True

primes = [x for x in range(50) if is_prime(x)]
print(f"Primes below 50: {primes}")
print(f"Sum: {sum(primes)}")
"""

sbx = Sandbox.create(template="python", timeout=300)
try:
    sbx.files.write("/tmp/code.py", code)
    result = sbx.commands.run("python3 /tmp/code.py", timeout=30)
    print(result.stdout)
finally:
    sbx.kill()

​
Expected Output

In demo mode: 
=======================================================
CrewAI + Declaw Sandbox Example
=======================================================
No OPENAI_API_KEY found -- running demo mode.

--- Tool Definition ---
from crewai.tools import tool
from declaw import Sandbox

@tool("Execute Python Code")
def execute_python(code: str) -> str:
    ...

--- Running Code Directly in Declaw Sandbox ---
Code:
def is_prime(n): ...

stdout: Primes below 50: [2, 3, 5, 7, 11, 13, 17, 19, 23, 29, 31, 37, 41, 43, 47]
Sum: 328
stderr:
exit_code: 0

Sandbox cleaned up.

=======================================================
Done!
=======================================================