Repolode
Submit
AutomatosAI

AutomatosAI/automatos-ai

Self-hosted AI operating system that runs a workforce of AI agents on your own machine, with any model, 1,000+ tool integrations and a Knowledge Graph.

View on GitHubWebsite
Stars
55
Forks
25
Open issues
74
Last commit
8 hours ago

What is it?

Automatos AI is an open-source platform for organizing AI agents as a team. Users give work to specialised agents through a single chat interface called Auto, which routes requests to the right agent, tool or Playbook, and the agents produce Deliverables such as reports, code and analysis. The local edition runs with docker compose and includes Postgres with pgvector, Redis and MinIO, with no login required.

Models can be connected through OpenRouter, direct provider keys for OpenAI, Anthropic, DeepSeek and others, or NVIDIA's hosted open models. Agents can also be backed by a user's own Claude Code sessions through a host process on the machine that holds the repositories, with approvals surfaced as cards in the board. Tool integrations are provided through the Composio catalogue and require a separate Composio API key.

The platform includes a command center with a ticket board and calendar, Playbooks and Missions for multi-step automation, RAG over uploaded documents, a Knowledge Graph built from them, and per-call cost analytics by provider and agent.

Key features

  • 01Auto chat interface that routes messages to the right agent, tool or Playbook.
  • 02Self-hosted local edition started with docker compose, with no login required.
  • 03Connects to OpenRouter, direct provider keys, and NVIDIA's hosted open models.
  • 04Session mode lets your own Claude Code sessions act as agents.
  • 05Over 1,000 tool integrations through Composio, with your own API key.
  • 06RAG over uploaded documents plus a Knowledge Graph built from them.
  • 07Per-call cost tracking by provider, agent and route.

Use cases

  • Running a team of AI agents for research, writing and analysis on your own infrastructure.
  • Connecting agents to tools like GitHub, Slack or Jira without writing per-tool glue code.
  • Letting a Claude Code subscription drive repository work under human approval.
  • Tracking and comparing LLM spend across providers and agents.

Quick start

  1. 1

    Clone the repository

    git clone --depth 1 https://github.com/AutomatosAI/automatos-ai.git && cd automatos-ai
  2. 2

    Start the local edition (writes .env with random secrets, then open http://localhost:3000)

    make up

Things to know

  • Runs a multi-service Docker stack (Postgres with pgvector, Redis, MinIO), so it needs a machine with Docker and enough resources.
  • You must bring your own model API keys, and the 1,000+ tool integrations need your own Composio key.
  • Session mode with Claude Code is local-only by design.

Topics

#ai-agents#multi-agent#self-hosted#llm#rag#knowledge-graph#docker#fastapi#agent-orchestration#ai-operating-system

FAQ

What is automatos-ai?
Self-hosted AI operating system that runs a workforce of AI agents on your own machine, with any model, 1,000+ tool integrations and a Knowledge Graph. Automatos AI is an open-source platform for organizing AI agents as a team. Users give work to specialised agents through a single chat interface called Auto, which routes requests to the right agent, tool or Playbook, and the agents produce Deliverables such as reports, code and analysis. The local edition runs with docker compose and includes Postgres with pgvector, Redis and MinIO, with no login required.
Is automatos-ai free to use? What license does it have?
automatos-ai is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
Is automatos-ai still maintained?
Actively maintained. The last commit to automatos-ai was on Oct 21, 2026.
What can I use automatos-ai for?
Running a team of AI agents for research, writing and analysis on your own infrastructure. Connecting agents to tools like GitHub, Slack or Jira without writing per-tool glue code. Letting a Claude Code subscription drive repository work under human approval. Tracking and comparing LLM spend across providers and agents.
How do I install automatos-ai?
Clone the repository: git clone --depth 1 https://github.com/AutomatosAI/automatos-ai.git && cd automatos-ai Start the local edition (writes .env with random secrets, then open http://localhost:3000): make up
Where can I get automatos-ai?
The source code and installation instructions are on GitHub: https://github.com/AutomatosAI/automatos-ai

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects