ai-parrot-tools is a collection of tools and toolkits for AI-Parrot agents. Each tool wraps an external API or service behind a unified interface that agents can discover and invoke automatically.
pip install ai-parrot-toolsInstall only the extras you need:
# Individual extras
pip install ai-parrot-tools[aws]
pip install ai-parrot-tools[slack]
pip install ai-parrot-tools[jira]
pip install ai-parrot-tools[google]
pip install ai-parrot-tools[finance]
pip install ai-parrot-tools[db]
# Everything
pip install ai-parrot-tools[all]| Extra | Description |
|---|---|
jira |
Jira issue tracking |
slack |
Slack messaging |
aws |
AWS services (EC2, S3, RDS, Lambda, ECS, IAM, etc.) |
docker |
Docker container management |
git |
Git repository operations |
analysis |
Data analysis with pandas, numpy, autoviz |
excel |
Excel and ODF spreadsheet support |
sandbox |
Sandboxed code execution via Docker |
codeinterpreter |
Interactive code interpreter |
pulumi |
Pulumi infrastructure-as-code |
sitesearch |
Website crawling and search |
office365 |
Microsoft 365 / Graph API integration |
scraping |
Web scraping with Selenium |
finance |
Financial analysis (TA-Lib, yfinance, Alpaca) |
db |
Database querying via QuerySource |
flowtask |
FlowTask workflow integration |
google |
Google APIs (Search, Maps, Routes) |
arxiv |
arXiv paper search |
wikipedia |
Wikipedia lookups |
weather |
OpenWeatherMap |
messaging |
MQTT and IMAP messaging |
from parrot.bots import Agent
agent = Agent(
name="assistant",
instructions="You are a helpful assistant.",
tools=["google_search", "weather", "calculator"],
)
response = await agent.chat("What's the weather in New York?")Tools are registered by name. Pass a list of tool names to any AI-Parrot agent and they are loaded on demand.
Use the @tool decorator for simple functions:
from parrot.tools import tool
@tool
def get_stock_price(symbol: str) -> str:
"""Get the current stock price for a ticker symbol."""
...Use AbstractToolkit for multi-tool collections:
from parrot.tools import AbstractToolkit
class MyToolkit(AbstractToolkit):
name = "my_toolkit"
description = "A collection of related tools."
def get_tools(self):
...- Python >= 3.11
- ai-parrot >= 0.23.18
MIT