Pydantic AI-based multi-agent framework with hierarchical task delegation, round-based evaluation, and workspace CLI
Project description
A multi-agent orchestration framework for LLM-powered workflows. MixSeek enables hierarchical task delegation with Leader and Member agents, round-based evaluation, and parallel team execution.
Features
- Multi-agent orchestration with Leader/Member hierarchy
- Support for multiple LLM providers (Google, OpenAI, Anthropic, xAI)
- Round-based evaluation with customizable judgment criteria
- Parallel team execution with leaderboard ranking
- Streamlit-based web UI for execution monitoring
- CLI tools for workspace management and agent execution
Installation
As a CLI Tool (Recommended)
Install globally as a command-line tool using uv:
# Install as a CLI tool (requires uv)
uv tool install mixseek-core
# Or install from source
git clone https://github.com/mixseek/mixseek-core.git
cd mixseek-core
uv tool install .
After installation, mixseek commands will be available globally:
mixseek --version
As a Python Package
Install as a library for programmatic use:
# Using pip
pip install mixseek-core
# Using uv
uv pip install mixseek-core
# Add to project dependencies (uv)
uv add mixseek-core
For development setup, see Development Guide.
Quick Start
Set Environment Variables
Choose one LLM provider and set the corresponding API key:
# Workspace directory (required)
export MIXSEEK_WORKSPACE=$HOME/mixseek-workspace
# Google Gemini (default in sample configs)
export GOOGLE_API_KEY=your-api-key
# OpenAI
export OPENAI_API_KEY=your-api-key
# Anthropic
export ANTHROPIC_API_KEY=your-api-key
# xAI (Grok)
export GROK_API_KEY=your-api-key
Initialize Workspace
mixseek init
This creates sample configuration files in your workspace:
configs/search_news.toml- Simple orchestratorconfigs/search_news_multi_perspective.toml- Multi-team orchestratorconfigs/agents/- Team configurations
Run Orchestration
# Execute with orchestrator config
mixseek exec "Search for the latest AI news" \
--config $MIXSEEK_WORKSPACE/configs/search_news.toml
# Run single team
mixseek team "Analyze this topic" \
--config $MIXSEEK_WORKSPACE/configs/agents/team_general_researcher.toml
# Launch web UI
mixseek ui
Supported LLM Providers
| Provider | Model Format | Environment Variable |
|---|---|---|
| Google Gemini | google-gla:gemini-2.5-flash |
GOOGLE_API_KEY |
| Google Vertex AI | google-vertex:gemini-2.5-flash |
GOOGLE_APPLICATION_CREDENTIALS |
| OpenAI | openai:gpt-4o |
OPENAI_API_KEY |
| Anthropic | anthropic:claude-sonnet-4-5-20250929 |
ANTHROPIC_API_KEY |
| xAI | xai:grok-3 |
GROK_API_KEY |
Documentation
Full documentation: https://mixseek.github.io/mixseek-core/
- Getting Started (Basic) - 5-minute quick start guide
- Getting Started (Advanced) - Multi-perspective search and customization
- Quick Start Guide - Detailed setup instructions
- Member Agents Guide - Agent types and configuration
- Team Guide - Team execution and delegation
- Orchestrator Guide - Multi-team parallel execution
- Configuration Reference - All configuration options
- Docker Setup - Container-based development
Contributing
Contributions are welcome! Please read our contributing guidelines before submitting pull requests.
License
Apache License 2.0 - see LICENSE for details.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file mixseek_core-0.1.0a15.tar.gz.
File metadata
- Download URL: mixseek_core-0.1.0a15.tar.gz
- Upload date:
- Size: 215.1 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: uv/0.11.7 {"installer":{"name":"uv","version":"0.11.7","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"24.04","id":"noble","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":true}
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
9e8a78ad3c50c527cdb982312a144a240a0266570ca23a09f00838930e7f1d26
|
|
| MD5 |
4a9a09b9aff4e28c4a8e32c98b94dbd0
|
|
| BLAKE2b-256 |
278df0ad6034710c80485750db36752a9931c3f00e921439f89573be21914eee
|
File details
Details for the file mixseek_core-0.1.0a15-py3-none-any.whl.
File metadata
- Download URL: mixseek_core-0.1.0a15-py3-none-any.whl
- Upload date:
- Size: 312.2 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: uv/0.11.7 {"installer":{"name":"uv","version":"0.11.7","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"Ubuntu","version":"24.04","id":"noble","libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":true}
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
3781c72272a4a26f3cd51271b0d16b14bc439f466fa0cafe02a63a6ed14ff788
|
|
| MD5 |
552e3785a86e2ac6c5772a58dfbd4465
|
|
| BLAKE2b-256 |
2647985894a5686e3ad871c701c0c81761afdca5df37624e9266c9acea6224e1
|