---
slug: "omnigent"
name: "Omnigent"
effect: "Orchestrates Claude Code, Codex, Cursor and other agents in one session, with policies and sandboxing"
category: "agent-framework"
tier: "rising"
url: "https://omnigent.ai"
github: "omnigent-ai/omnigent"
pricing: "open-source"
license: "Apache-2.0"
tags: ["meta-harness", "orchestration", "multi-agent", "policies", "sandbox", "claude-code", "codex"]
platforms: ["claude-code", "codex", "cursor", "macos", "linux", "windows", "web", "docker"]
install_kind: "cli"
date_added: "2026-10-16"
rank: 12
rising_score: 56
stars: 10841
star_velocity: 85.4
velocity_basis: "since creation"
created_at: "2026-06-11T12:18:13Z"
pushed_at: "2026-10-16T11:44:30Z"
signals_status: "ok"
signals_fetched_at: "2026-10-16T11:45:34.470Z"
page: "https://agenttooldrop.com/tools/omnigent"
---

# Omnigent

Orchestrates Claude Code, Codex, Cursor and other agents in one session, with policies and sandboxing

Category: Agent frameworks. Tier: new and rising, rank 12 by new and rising score (56 of 100). Pricing: Open source. License: Apache-2.0.

Platforms: claude-code, codex, cursor, macos, linux, windows, web, docker.

## Why you would want it

Omnigent is a meta-harness that puts a common orchestration layer over Claude Code, Codex, Cursor, OpenCode, Hermes, Pi and agents you write yourself, so you can swap or combine harnesses without rewriting. Agents are short YAML files with a prompt, local Python functions, MCP servers and sub-agents, and several agents can share one session, for example one reviewing another's work. Policies can pause for your approval before risky actions, cap spend or limit which tools an agent reaches, at server, agent or session level. Sessions can be followed from the terminal, a browser or a phone, and can run in cloud sandboxes such as Modal, Daytona or E2B.

## Best for

- Running Claude Code and Codex in one session
- Approval and spend policies across agents
- Sharing live agent sessions with a team

## Use it when

You want one layer to run and supervise several coding agents, with approval policies and sessions you can share or pick up from another device.

## Skip it if

The README badges it alpha and says anonymized telemetry is collected by default, with an opt-out documented. It picks up model credentials from your environment, runs agents that execute commands on your machine, and lets a co-attached teammate's messages execute there too. On native Windows the Job Object backend does not isolate the filesystem or network.

## Editor note

Needs Python 3.12+. The README's headline installer is a curl-pipe-shell script, also listed alongside Homebrew; the snippet shown here is the README's manual route. Linux needs bubblewrap for the native terminal wrappers.

## Install

Install with uv or pip:

```sh
uv tool install omnigent        # or: pip install "omnigent"
```

Source: https://github.com/omnigent-ai/omnigent#1-install

## Live signals

Fetched 2026-10-16T11:45:34.470Z (status: ok).

- Repository: omnigent-ai/omnigent
- Repository created: 2026-06-11
- Stars: 10,841 (GitHub's count; it cannot be authenticated)
- Stars per day, since repo creation: 85.4 (an average over the repo's life, not a growth rate)
- Forks: 1,755
- Open issues and PRs: 2,143
- License (GitHub): Apache-2.0
- Last push: 2026-10-16
- Latest release: v0.18.0 (2026-10-13)

## Links

- Website: https://omnigent.ai
- GitHub: https://github.com/omnigent-ai/omnigent
- Install instructions verified at: https://github.com/omnigent-ai/omnigent#1-install
- Source for this entry: https://github.com/omnigent-ai/omnigent#readme
- AgentToolDrop page: https://agenttooldrop.com/tools/omnigent
- Full catalog (JSON): https://agenttooldrop.com/tools.json