Octop

Octop brings AI agent deployment to your own infrastructure. Instead of scattering your work across cloud services, you get a unified platform running on your machine with a web dashboard, CLI, and messaging integrations. The system supports multiple LLM providers—use OpenAI, Qwen, Ollama, or others—and lets each team member maintain their own agent roster with personality templates and specialized skills.

Highlights

• Run agents locally with full control over data, credentials, and conversation history
• Multi-user workspace where each person manages their own AI agents with switchable personalities
• Messaging integrations connect to Discord, Slack, DingTalk, Feishu, and your terminal
• Support for multiple LLM backends including open-source models via Ollama
• Browser automation, IDE integration via Agent Protocol, and tool creation for extended capabilities

Octop is built for teams and individuals who need AI assistance without compromising privacy or sending sensitive information to cloud services. It's ideal for organizations handling confidential data, developers wanting local AI for coding tasks, or anyone building agent-based workflows. Completely free and open-source under MIT license, so you can deploy, modify, and scale it however you need.