Skip to main content

Use Composio to get an array of tools with your CrewAI agent.

Project description

Composio <> CrewAI

CrewAI Logo

Composio enables CrewAI agents to connect with numerous tools, making it easy for these agents to interact with external applications seamlessly.

Installation and Setup

Start by installing Composio CrewAI and connecting your GitHub account to enable GitHub functionalities for your agents.

pip install composio_crewai
composio-cli add github  # Connect your GitHub account
composio-cli show-apps   # Check available applications

Goal

Automate GitHub interactions, such as starring a repository, using natural language commands through a CrewAI Agent.

Step-by-Step Guide

1. Import Base Packages

Prepare your environment by importing the necessary packages for CrewAI and setting up your language model.

from crewai import Agent, Task
from langchain_openai import ChatOpenAI

# Initialize the language model with your OpenAI API key
llm = ChatOpenAI(openai_api_key="sk-<OPENAI KEY>")

2. Fetch Tools via Composio

Initialize the Composio toolset for interacting with GitHub.

from composio_crewai import ComposioToolset, Action, App

# Initialize the toolset with GitHub application
tools = ComposioToolset(apps=[App.GITHUB])

3. Execute the Agent

Configure and execute the CrewAI agent to perform GitHub actions.

crewai_agent = Agent(
    role='Github Agent',
    goal="You take action on Github using Github APIs",
    backstory="You are an AI agent responsible for taking actions on Github on users' behalf using Github APIs",
    verbose=True,
    tools=tools,
    llm=llm
)

# Define the task
task = Task(
    description="Star a repo SamparkAI/docs on GitHub",
    agent=crewai_agent,
    expected_output="if the star happened"
)

# Execute the task
task.execute()

4. Check Response

Verify the agent's actions and responses to ensure the task was completed successfully.

> Entering new CrewAgentExecutor chain...
> I need to star the repository "SamparkAI/docs" on GitHub.
> {'execution_details': {'executed': True}, 'response_data': ''}
> Finished chain.

Advanced Configuration

  • Filter Specific Actions: Limit the actions an agent can perform for enhanced security and operational focus.
toolsGithubCreateIssue = ComposioToolset(actions=[Action.GITHUB_CREATE_ISSUE])
  • Filter Specific Apps: Restrict the agent's access to certain applications for streamlined operations.
toolsAsanaGithub = ComposioToolset(apps=[App.ASANA, App.GITHUB])

Project details


Release history Release notifications | RSS feed

This version

0.3.7

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

composio_crewai-0.3.7.tar.gz (2.6 kB view details)

Uploaded Source

Built Distribution

composio_crewai-0.3.7-py3-none-any.whl (2.7 kB view details)

Uploaded Python 3

File details

Details for the file composio_crewai-0.3.7.tar.gz.

File metadata

  • Download URL: composio_crewai-0.3.7.tar.gz
  • Upload date:
  • Size: 2.6 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.0.0 CPython/3.12.3

File hashes

Hashes for composio_crewai-0.3.7.tar.gz
Algorithm Hash digest
SHA256 a5b2d40b3e3a95672cec9e5cad0c3480fa42a5c921d3f62d0e33a83e2b1cd416
MD5 3000a8141a1c9e5035060c693edbc64f
BLAKE2b-256 fb9ef26068c69acb484116a926ad497d749b83379f7a0e0544bf4074f3f3391e

See more details on using hashes here.

File details

Details for the file composio_crewai-0.3.7-py3-none-any.whl.

File metadata

File hashes

Hashes for composio_crewai-0.3.7-py3-none-any.whl
Algorithm Hash digest
SHA256 f2d2b49c90e6dca14944eee1a329eac05227a58f826b163a9e3ecee9ab6b55a9
MD5 1ca80673f9346836d645735b1770c4eb
BLAKE2b-256 d9aec12fd2323454ecc14f1c83daaee0ea5c720319d520d5a869101558186ac6

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page