Skip to main content

The Conductor for Your AI Orchestra

Project description

Glee

The Conductor for Your AI Orchestra

An orchestration layer for AI coding agents with shared memory, context injection, and multi-agent collaboration.

Quick Start

# Install
uv tool install glee --python 3.13
# or: pipx install glee

# Upgrade
uv tool upgrade glee --python 3.13

# Initialize project (registers MCP server for Claude Code)
glee init

# Connect agents
glee connect claude --role coder --domain backend,api
glee connect codex --role reviewer --focus security,performance

# View status
glee status

# Run review
glee review src/main.py
glee review git:changes          # Review uncommitted changes
glee review git:staged           # Review staged changes

Features

  • MCP Integration: glee init registers Glee as an MCP server - Claude Code gets glee_* tools automatically
  • Multiple Coders: Different agents for different domains (backend, frontend, infra)
  • Multiple Reviewers: Get diverse perspectives (security, performance, architecture)
  • Parallel Execution: Reviews run concurrently for speed
  • Flexible Review Targets: Review files, directories, git changes, or natural descriptions

Claude Code Integration

After running glee init, restart Claude Code. You'll have these MCP tools:

  • glee_status - Show project status and connected agents
  • glee_review - Run multi-agent review on any target
  • glee_connect - Connect an agent to the project
  • glee_disconnect - Disconnect an agent
# In Claude Code, you can now say:
"Use glee_review to review the uncommitted changes"
"Connect codex as a security reviewer using glee"

CLI Commands

glee init                         # Initialize project + register MCP server
glee status                       # Show global and project status
glee connect <cmd> --role <role>  # Connect an agent
glee disconnect <name>            # Disconnect an agent
glee agents                       # List available agents
glee review [target]              # Run multi-reviewer workflow
glee test-agent <cmd>             # Test an agent
glee mcp                          # Run MCP server (used by Claude Code)

Review Targets

glee review src/api/              # Review a directory
glee review src/main.py           # Review a file
glee review git:changes           # Review uncommitted changes
glee review git:staged            # Review staged changes
glee review "the auth module"     # Natural description

Configuration

# .glee/config.yml
project:
  id: 550e8400-e29b-41d4-a716-446655440000
  name: my-app
  path: /Users/yumin/ventures/my-app

agents:
  - name: claude-a1b2c3
    command: claude
    role: coder
    domain: [backend, api]
    priority: 1

  - name: codex-d4e5f6
    command: codex
    role: reviewer
    focus: [security, performance]

dispatch:
  coder: first      # first | random | round-robin
  reviewer: all     # all | first | random

How It Works

glee init
    ├── Creates .glee/config.yml
    └── Creates .mcp.json (MCP server registration)

claude (start in project)
    └── Reads .mcp.json
        └── Spawns `glee mcp` as MCP server
            └── Claude now has glee_* tools

Documentation

See docs/PRD.md for full documentation.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

glee-0.2.15.tar.gz (184.9 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

glee-0.2.15-py3-none-any.whl (41.0 kB view details)

Uploaded Python 3

File details

Details for the file glee-0.2.15.tar.gz.

File metadata

  • Download URL: glee-0.2.15.tar.gz
  • Upload date:
  • Size: 184.9 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.9.17 {"installer":{"name":"uv","version":"0.9.17","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

File hashes

Hashes for glee-0.2.15.tar.gz
Algorithm Hash digest
SHA256 4284d1b29054d572f1832edfcbc69e9997ca68b42434ac270d0602b3d0a3ab42
MD5 d9fe2ac85a28ebf66e61e4a8955dd51a
BLAKE2b-256 89abd4e661aac239f1b957cf4aeab530445ea67108318ed99ed09b9006e88c80

See more details on using hashes here.

File details

Details for the file glee-0.2.15-py3-none-any.whl.

File metadata

  • Download URL: glee-0.2.15-py3-none-any.whl
  • Upload date:
  • Size: 41.0 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.9.17 {"installer":{"name":"uv","version":"0.9.17","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

File hashes

Hashes for glee-0.2.15-py3-none-any.whl
Algorithm Hash digest
SHA256 42ffae1861c9c5805ddae258e689367d721680064961c65852b12739a107e5b7
MD5 7929092dadd37d5faac86f090b00643e
BLAKE2b-256 403a2e446ad5e097a550fde2c9bec8aa8a07faff1cc72f5695473249fe1ee44f

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page