Back to directory

Octop

A self-hosted, multi-user, multi-agent AI assistant that runs entirely on a user's own machine, aiming to give teams, families and individuals an independent, collaborative agent environment.

Octop

Octop is reachable through a web dashboard, Feishu, DingTalk, QQ, WeChat, Telegram, Discord, WeCom, or programmatic HTTP, SSE or WebSocket access, with capabilities extended by an expert library, OAuth and MCP connectors, and ACP integration for IDE workflows.

It offers a shared multi-user expert team with an expert library and market, expert sharing so teammates reuse skill and sub-agent pools, sixteen MBTI-style personality templates, a beta coordinator for scheduling multiple experts on multi-step work, JWT multi-user isolation with tool approval and shell guardrails, pluggable workspace backends (local disk, Docker sandbox, PostgreSQL, or COS/S3), a portable memory system, retrieval-augmented generation over shared knowledge bases, third-party plugins, and bidirectional ACP so it can delegate to OpenCode or Claude Code with permission gates.

It is written in Python and licensed under AGPL-3.0.

More in Agent frameworks

See all 902
  • Recurse

    recurse.run

    A serverless harness that lets a coding agent design, test, and deploy its own specialist sub-agents as tools, MCP servers, or bots, defined by a schema-validated manifest.

  • ProofForge

    shulgin.is-a.dev

    An AI-agent pipeline that decomposes a mathematics problem, proves each piece, and formalizes the result in Lean 4 and Mathlib until the Lean kernel accepts it, with six of its proofs already merged into Google DeepMind's formal-conjectures repository.

  • local-coder

    github.com

    Builds a local coding-agent team on top of Ollama and OpenCode by detecting the machine's hardware, recommending models for seven specialist roles, and testing whether each assigned model can actually use tools, edit a repository and pass verification.