harnessrouter vs omnigent
Self-hosted Apache-2.0 reference implementation of the Unified Harness Protocol: one OpenAI-Responses-style API running Codex, Claude Code, Hermes, Pi and DSH with sessions, streaming, files, traces. — versus — Open-source meta-harness over Claude Code, Codex, Cursor, OpenCode, Hermes, Pi and custom agents: swap harnesses without rewriting, enforce policies and sandboxes, follow sessions from any device.
Both are meta-layers over Claude Code, Codex, Hermes, Pi and DSH that let you swap the harness; omnigent is a developer-facing meta-harness, harnessrouter is a self-hosted API service with sessions, files and a protocol spec.
| harnessrouter | omnigent | |
|---|---|---|
| Stars | 1.1k | 9.8k |
| Forks | 112 | 1.5k |
| Language | Python | Python |
| License | Apache-2.0 | Apache-2.0 |
| Last activity | yesterday | yesterday |
| Topics | orchestration, agents, gateway | agents, orchestration, coding |
| Curated connections | 5 | 9 |
harnessrouter — the curator's take
Use it when your *product* needs to run coding-agent harnesses as a backend service and you don't want an integration per harness: one Docker image ships Console + Gateway + Runner, installs the harness CLIs you enable, and exposes runs/sessions/artifacts through UHP so you can swap Codex for Claude Code (or a cheaper harness×model pair — their benchmarks show 90%+ cost spread) without touching your app. Bring your own provider keys. It is a server for embedding harnesses, not a desktop cockpit for a developer (see agent-orchestrator, helmor) and not a model gateway (put litellm/omniroute underneath). Community Edition of a commercial product; check what stays in the paid tier before betting on it.
omnigent — the curator's take
Use when you run several coding harnesses and want one control plane: mix agents in a session, gate risky actions with policies, run in Modal/Daytona/E2B/K8s sandboxes, and pick a session up from phone or browser. Team sharing (co-drive, fork) is a real differentiator. NOT if you live in one CLI on one laptop — it's a server + desktop app stack, and it's alpha; expect churn. Apache-2.0.