Omnigent
ActiveDescription
Open-source AI agent framework and meta-orchestrator that unifies Claude Code, Codex, Cursor, Pi, and custom agents with policy enforcement, sandboxing, and real-time collaboration.
Key Features
- Multi-agent orchestration — Unifies Claude Code, Codex, Cursor, Pi, and custom agents with seamless framework switching
- Policy and sandbox governance — Centrally defined and enforced access policies, resource limits, and sandbox boundaries
- Real-time collaboration — Multi-user real-time collaborative debugging of agent tasks from any device
- Framework agnostic — Write agents once, switch between agent frameworks without rewriting
- Meta-orchestrator architecture — Unified interface on top of agent frameworks for cross-cutting concerns
- Extensible hook system — Custom policy engines, audit logs, and callback hooks
Use Cases
💡 Unify multiple AI coding agents (Claude Code, Codex, Cursor) in enterprise environments
💡 Enforce uniform access control and security policies across multiple agents
💡 Real-time cross-device collaboration on AI agent task development and debugging
💡 Migrate seamlessly between agent frameworks without rewriting business logic
💡 Establish audit logs and compliance tracking for AI agent teams
Strengths & Limitations
✅ Strengths
- • Actively maintained, recent updates
- • High community interest (9.6k stars)
- • Permissive open-source license (Apache-2.0)
⚠️ Limitations
- • High issue backlog (1.2k open issues)
Categories
Quick Start
pip install omnigent
# Initialize config
omnigent init
# Register agents
cat > agents.yaml <<EOF
agents:
- name: claude-code
type: claude-code
api_key: $ANTHROPIC_API_KEY
- name: codex
type: codex
api_key: $OPENAI_API_KEY
EOF
# Start orchestrator
omnigent serve --config agents.yaml --port 7700