📖 Using Composio With PraisonAI

Composio enables your PraisonAI agents to connect with many tools!

Goal: Star a repository on GitHub with natural language & PraisonAI Agent

Install Packages & Connect a Tool

These commands prepare your environment for seamless interaction between PraisonAI and GitHub.

Run command

$ pip install PraisonAI -q
> pip install composio-praisonai
> # login to composio
> composio login
> # Connect your GitHub using command below, so agents can use it. 
> composio add github
> # Check all different apps which you can connect with
> composio apps

Goal: Use PraisonAI Agent to Interact with Github using Composio

Import Base Packages

Prepare your environment by initializing necessary imports from PraisonAI and setting up your client.

Default Imports

1 import os
2 import yaml
3 from praisonai import PraisonAI
4 
5 from composio_praisonai import Action, ComposioToolSet

Write the Praison-supported Composio Tools in `tools.py` file.

This step involves fetching and integrating GitHub tools provided by Composio, and writing them in PraisonAI supported Format, returning the name of tools in a format, that should be added to agents.yml file.

Write the tools

1 composio_toolset = ComposioToolSet()
2 tools = composio_toolset.get_tools(
3     actions=[Action.GITHUB_ACTIVITY_STAR_REPO_FOR_AUTHENTICATED_USER]
4 )
5 tool_section_str = composio_toolset.get_tools_section(tools)
6 print(tool_section_str)

Define the `agents.yml` either in a separate file, or in your script.

This step involves configuring and executing the agent to carry out actions, such as starring a GitHub repository.

Define agent.yml

1 agent_yaml = """
2 framework: "crewai"
3 topic: "Github Management"
4 
5 roles:
6   developer:
7     role: "Developer"
8     goal: "An expert programmer"
9     backstory: "A developer exploring new codebases and having certain tools available to execute different tasks."
10     tasks:
11       star_github:
12         description: "Star a repo composiohq/composio on GitHub"
13         expected_output: "Response whether the task was executed."
14 """ + tool_section_str
15 
16 print(agent_yaml)

Run the PraisonAI Agents to execute the goal/task.

Here you initialize PraisonAI class, and execute.