# omnara-ai/omnara

The open-source alternative to Claude Managed Agents

Repository: https://github.com/omnara-ai/omnara
Canonical: https://ross.abutalabs.com/products/omnara
Homepage: https://omnara.com
Language: Go
License: Apache-2.0
License Family: permissive
Topics: agent-infrastructure, managed-agents, durable-agents, coding-agents
Last push: 2026-08-26T20:24:26+00:00

## Health v2 (maintenance only)
Score: 85/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 99, longevity 30
- inputs: {"age_days": 421, "days_push": 7, "days_rel": 8, "gap_med": 0, "n_releases_24m": 30}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2765, forks 212 (observed 2026-08-28T04:07:18.612448+00:00)

## What it is
Omnara is an open-source platform for running managed, durable AI agents, handling execution and state while you bring your own models, tools, machines, and interaction surfaces. It offers a REST API, CLI, and SDK, with a hosted cloud option or self-hosting under Apache 2.0.

## Use cases
- run production-grade AI agents with durable state
- self-host an alternative to Claude managed agents
- build customer-facing agents via an API
- recover agents from crashes and restarts automatically
- connect coding agents to sandboxes and my own machines
- query agent history in Postgres for evals and training data
- manage agent access with RBAC across a team
- chat with agents through Slack or a dashboard

## When to choose
- you need durable, crash-recoverable agents with queryable state in Postgres
- you want to bring your own models via OpenRouter, LiteLLM, or Ollama
- you need RBAC and multi-user agent management for teams
- you want a self-hosted, Apache-2.0 alternative to managed agent platforms
- you need to attach or swap execution machines (sandboxes, VMs, laptops) while agents run

## When to avoid
- you only need a simple single-script agent loop without state management
- you want a fully managed vendor solution and prefer not to run infrastructure
- your stack requires agent frameworks outside the supported OpenAI/Anthropic API formats
- you need a lightweight library rather than a platform with Postgres and services

## Facets
- artifact type: service
- maturity: active
- function: agent-framework, llm-inference, mcp, api-framework, self-hosted, authorization, cli, sdk
- domain: large-language-models, developer-tools, self-hosted
- platform: self-hosted, cli, go, cloud
- tags: managed-agents, durable-agents, agent-infrastructure, coding-agents, agent-orchestration, rbac, postgres-state, sandboxed-execution, openrouter, slack-connector, ai-agents, automation, docker, web-server

## Member repositories
- omnara-ai/omnara (main) score 85

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:18.612448+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T08:16:46.168716+00:00, confidence not recorded.
  - readme: https://github.com/omnara-ai/omnara (fetched 2026-08-28T04:07:18.612448+00:00, sha 9b4f7cee02dc)
  - homepage: https://omnara.com (fetched 2026-08-29T09:56:18.741714+00:00, sha 04a6970fae77)
- Data as of 2026-08-30T08:39:29.467469+00:00.