AI Orchestration Tool for Claude Code - Multi-agent orchestration for complex development tasks
Project description
Ensemble
"One task. Many minds. One result."
AI Orchestration Tool for Claude Code.
Overview
Ensemble is an AI orchestration system that combines the best practices from:
- shogun - Autonomous AI collaboration with tmux parallel execution
- takt - Workflow enforcement with quality gates
- Boris's practices - Effective use of skills, subagents, CLAUDE.md, and hooks
Features
- Autonomous AI Coordination: One instruction triggers multiple AI agents working together
- Flexible Execution Patterns:
- Pattern A: Simple tasks via subagent
- Pattern B: Medium tasks via tmux parallel panes
- Pattern C: Large tasks via git worktree separation
- Parallel Review: Architecture + Security reviews run in parallel
- Self-Improvement: Automatic learning and CLAUDE.md updates
- Compaction Recovery: Built-in protocol to prevent role amnesia
Installation
Using uv (recommended)
# Install globally
uv tool install ensemble-claude
# Or add to your project
uv add ensemble-claude
Using pip
pip install ensemble-claude
From source
git clone https://github.com/ChikaKakazu/ensemble.git
cd ensemble
# Using uv
uv pip install -e .
# Or using pip
pip install -e .
Quick Start
# 1. Initialize Ensemble in your project
ensemble init
# 2. Launch the tmux session with Conductor + Dispatch
ensemble launch
# 3. Run a task (in the Conductor pane)
/go implement user authentication
# Light workflow (minimal cost)
/go-light fix typo in README
CLI Commands
| Command | Description |
|---|---|
ensemble init |
Initialize Ensemble in current project |
ensemble init --full |
Also copy agent/command definitions locally |
ensemble launch |
Start tmux session with Conductor + Dispatch |
ensemble launch --no-attach |
Start session without attaching |
ensemble --version |
Show version |
Requirements
- Python 3.10+
- Claude Code CLI (
claudecommand available) - tmux
- git 2.20+ (for worktree support)
- Claude Max plan recommended (for parallel execution)
Agent Architecture
┌─────────────┐
│ Conductor │ ← Orchestrator (planning, judgment, delegation)
└──────┬──────┘
│
┌────┴────┐
▼ ▼
┌────────┐ ┌──────────┐
│Dispatch│ │ Learner │
└───┬────┘ └──────────┘
│ ↑ Learning records
▼
┌─────────────────────────────┐
│ Reviewer / Security-Reviewer│ ← Parallel reviews
└─────────────────────────────┘
│
▼ (worktree mode)
┌──────────┐
│Integrator│ ← Merge & integrate
└──────────┘
License
MIT License - 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 ensemble_claude-0.3.1.tar.gz.
File metadata
- Download URL: ensemble_claude-0.3.1.tar.gz
- Upload date:
- Size: 58.8 kB
- Tags: Source
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/6.1.0 CPython/3.13.7
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
90786c755b924edf7a5bee909f3dc0785d46b82dab89a4e34617e554492adf00
|
|
| MD5 |
ac656d8cedd6401590bcacf772c93f2e
|
|
| BLAKE2b-256 |
4fd21abdd34d2af9458a1e93c11c4448da0f8e4d78eb962bdf0fa413b2798580
|
Provenance
The following attestation bundles were made for ensemble_claude-0.3.1.tar.gz:
Publisher:
publish.yml on ChikaKakazu/ensemble
-
Statement:
-
Statement type:
https://in-toto.io/Statement/v1 -
Predicate type:
https://docs.pypi.org/attestations/publish/v1 -
Subject name:
ensemble_claude-0.3.1.tar.gz -
Subject digest:
90786c755b924edf7a5bee909f3dc0785d46b82dab89a4e34617e554492adf00 - Sigstore transparency entry: 908796754
- Sigstore integration time:
-
Permalink:
ChikaKakazu/ensemble@08e3e2b419892acf677ca1926de4357f65d8f062 -
Branch / Tag:
refs/heads/deploy/production - Owner: https://github.com/ChikaKakazu
-
Access:
public
-
Token Issuer:
https://token.actions.githubusercontent.com -
Runner Environment:
github-hosted -
Publication workflow:
publish.yml@08e3e2b419892acf677ca1926de4357f65d8f062 -
Trigger Event:
push
-
Statement type:
File details
Details for the file ensemble_claude-0.3.1-py3-none-any.whl.
File metadata
- Download URL: ensemble_claude-0.3.1-py3-none-any.whl
- Upload date:
- Size: 69.9 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? Yes
- Uploaded via: twine/6.1.0 CPython/3.13.7
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
9972cc6126ea403a1a0e04eac38d3bac26e57cb9ff243ce8de904827ce81a0c9
|
|
| MD5 |
ca3d27baab22eeb812448bc28d7e14e4
|
|
| BLAKE2b-256 |
cb2f99e360e892e96dfb6ba2710cfebb77faab61a1c5593407c6cfa92dfc61a3
|
Provenance
The following attestation bundles were made for ensemble_claude-0.3.1-py3-none-any.whl:
Publisher:
publish.yml on ChikaKakazu/ensemble
-
Statement:
-
Statement type:
https://in-toto.io/Statement/v1 -
Predicate type:
https://docs.pypi.org/attestations/publish/v1 -
Subject name:
ensemble_claude-0.3.1-py3-none-any.whl -
Subject digest:
9972cc6126ea403a1a0e04eac38d3bac26e57cb9ff243ce8de904827ce81a0c9 - Sigstore transparency entry: 908796758
- Sigstore integration time:
-
Permalink:
ChikaKakazu/ensemble@08e3e2b419892acf677ca1926de4357f65d8f062 -
Branch / Tag:
refs/heads/deploy/production - Owner: https://github.com/ChikaKakazu
-
Access:
public
-
Token Issuer:
https://token.actions.githubusercontent.com -
Runner Environment:
github-hosted -
Publication workflow:
publish.yml@08e3e2b419892acf677ca1926de4357f65d8f062 -
Trigger Event:
push
-
Statement type: