Skip to main content

Council

Council: AI Agent Platform with Control Flow and Scalable Oversight

Supported Python versions License PyPI version

Welcome

Council is an open-source platform for building applications with Large Language Models (LLMs) - built in Python.

Council provides a unified interface for working with different LLM providers like OpenAI, Anthropic, Google and Ollama. The framework makes it easy to switch between providers while maintaining consistent interfaces and monitoring capabilities.

Council aims to provide enterprise-grade quality control and monitoring for LLM applications (contributions are welcome).

Key Features

  • 🧐 Unified LLM Interface: Consistent API across different LLM providers with built-in error handling and retries
  • 🔄 Provider Flexibility: Easy switching between LLM providers like OpenAI, Anthropic, Google Gemini, and local models via Groq and Ollama
  • 📊 Usage Monitoring: Built-in consumption tracking and monitoring capabilities
  • 🛠️ Configuration Management: Flexible configuration system for LLM parameters like temperature, max tokens etc.
  • 🔒 Error Handling: Robust error handling and retry mechanisms for production use

Key Concepts

LLM Interface

The core of Council is the LLM interface which provides a unified way to interact with different language model providers. This includes:

  • Flexible configuration options
  • Consistent message formatting across providers
  • Built-in retry mechanisms
  • Usage tracking and monitoring

Configuration

Council provides a robust configuration system that allows you to:

  • Set provider-specific parameters
  • Configure retry behavior
  • Control model parameters like temperature, max tokens etc.
  • Manage API credentials

Monitoring

Built-in monitoring capabilities help track:

  • Token usage and costs
  • Number of API calls
  • Response times

Quickstart

Installation

Install Council in one of multiple ways:

  1. (Recommended) Install with pip via Pypi: pip install council-ai
  2. Install with pip from git ref: pip install git+https://github.com/chain-ml/council.git@<branch_name>
  3. Install with pip from local copy:
    • Clone this repository
    • Navigate to local project root and install via pip install -e git+https://github.com/chain-ml/council.git@<branch_name>.

Uninstall with: pip uninstall council-ai

Current Stable Version

GitHub release (latest SemVer)

Setup

Set up your required API keys in a .env file (e.g. OpenAI). Refer to .env.example as an example.

Linter

Use make lint to verify your code.

Black

Use black . to automatically reformat files.

Documentation

A detailed documentation of Council can be found at council.dev.

Support

Please submit a GitHub issue should you need any help or reach out to the team via Discord.

Contributors

Council is a project under active development. We welcome all contributions, pull requests, feature requests or reported issues.

Metadata

Release files for council-ai 0.0.29

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for council-ai 0.0.29
File Size Uploaded Provenance
council_ai-0.0.29.tar.gz 92.8 kB Without provenance Details

Built distribution (wheel)

Table of built distributions (wheels) for council-ai 0.0.29
File Interpreter ABI Platform
council_ai-0.0.29-py3-none-any.whl Python 3 none any Details

Total release size: 254.3 kB

Release files / council_ai-0.0.29.tar.gz

Download URL council_ai-0.0.29.tar.gz
Size 92.8 kB
Tags Source
SHA-256 checksum
How to use checksums
84b3b066dbf0692732cd5a6faec9d9a8756c827576d94dadb9beb369d3f17830
BLAKE2b-256 checksum
How to use checksums
9c3661dc0c0685072fc2e71b13e0ef933de2d9fcc5fe07d58c237542919fc955
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.0.1 CPython/3.12.8

Release files / council_ai-0.0.29-py3-none-any.whl

Download URL council_ai-0.0.29-py3-none-any.whl
Size 161.5 kB
Tags Python 3
SHA-256 checksum
How to use checksums
3b9224f81bf7b5bea62bed003b170186ad9e024b23b84cfb6ade3158a8f857da
BLAKE2b-256 checksum
How to use checksums
b948b416de24d8787e841afcbeabe196c33b79e0158b7bf02b226e5d9ab91de0
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/6.0.1 CPython/3.12.8

Security for council-ai 0.0.29

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page