TaskWeaver
StaleDescription
TaskWeaver is Microsoft's open-source code-interpreter-style agent framework, suitable for data analysis and complex task automation.
Key Features
- Code-first agent framework: interprets user requests through code snippets with plugin-based execution
- Stateful execution preserving both chat history and code execution history including in-memory data
- Rich data structure support: works with DataFrames and complex Python data types instead of plain strings
- Customizable plugins: encapsulate domain-specific algorithms and orchestrate them for complex tasks
- Reflective execution: agent reflects on execution process and makes adjustments for better results
- Container-based code execution for security, with session management for multi-user data isolation
Use Cases
π‘ Complex data analytics tasks with DataFrame manipulation and multi-step reasoning
π‘ Prototyping AI-powered data analysis workflows with custom algorithm plugins
π‘ Multi-turn data exploration where the agent maintains state across interactions
π‘ Enterprise data analysis with domain-specific knowledge integration and security controls
Categories
Quick Start
git clone https://github.com/microsoft/TaskWeaver.git
cd TaskWeaver
pip install -r requirements.txt
# Configure your LLM endpoint in config.yaml
# Then run:
python -m taskweaver.app
# Or use Docker:
docker pull taskweaver/taskweaver:latest
docker run -p 8000:8000 taskweaver/taskweaver:latest