clowder-ai
Build AI teams, not just agents. Hard rails, soft power, shared mission.
Clowder AI is a self-hosted workspace for building persistent AI agent teams that collaborate across different model families including Claude, GPT, and Gemini. Unlike disposable chat sessions, it provides agents with stable identities, shared context, evidence-backed memory, and cross-model review capabilities. The platform handles agent-to-agent handoffs through @mentions, maintains work ownership visibility, and includes operational guardrails like review gates and worktree isolation. It connects to existing agent CLIs (Claude Code, Codex CLI, Gemini CLI, opencode) and adds a collaboration layer for long-term continuity.
Key Features
Use Cases
- 01Managing long-term software projects with multiple AI agents handling research, implementation, and review
- 02Cross-model validation where one LLM writes code and another independently reviews it
- 03Building institutional knowledge that persists across model upgrades and context resets
- 04Delegating complex multi-step tasks to agent teams with visible ownership and handoff tracking
- 05Collaborative AI development where agents from different families contribute specialized expertise
- 06Maintaining continuity in AI-assisted work without re-establishing context in every session
Related Agents
View morehermes-agent
The agent that grows with you
agency-agents
A complete AI agency at your fingertips - From frontend wizards to Reddit community ninjas, from whimsy injectors to reality checkers. Each agent is a specialized expert with personality, processes, and proven deliverables.
openinterpreter
A coding agent for open models like Kimi K3
cline
Autonomous coding agent as an SDK, IDE extension, or CLI assistant.
clowder-ai — FAQ
What is Clowder AI?+
Clowder AI is a self-hosted multi-agent workspace that enables AI agents from different model families (Claude, GPT, Gemini) to collaborate as a persistent team with shared memory, stable identities, and cross-model review. It adds a coordination layer on top of existing agent CLIs rather than replacing them.
How do I install Clowder AI?+
Install from source by cloning the GitHub repository, running pnpm install and pnpm build, configuring environment variables, and starting the server. Desktop installers are also available for Windows and macOS in the releases section. After installation, configure model provider credentials in the System Settings.
Which agent CLIs and models does Clowder AI support?+
Clowder AI supports Claude Code, Codex CLI (GPT), Antigravity CLI and Gemini CLI (Gemini models), and opencode for multiple providers. It works as a team layer above these agent CLIs, coordinating their collaboration while they handle individual tool use and commands.
What are the prerequisites for running Clowder AI?+
You need Node.js 20+, pnpm 9+, Git, and optionally Redis 7+ for enhanced memory features. You also need API access and credentials for the model providers you want to use (Anthropic, OpenAI, Google). The specific CLI tools for each model family must be installed separately.
Is Clowder AI free and open source?+
Yes, Clowder AI is released under the MIT license, allowing free use, modification, and distribution. However, you will incur costs from the underlying model providers (Claude, GPT, Gemini) based on your API usage.
How does memory and context persistence work?+
Clowder maintains shared evidence, Git history, task state, and approved memory that survives model upgrades and session resets. Observations remain as evidence until human confirmation promotes them to profile preferences, conventions, or system guards, ensuring memory earns trust over time.
How do I install clowder-ai?+
Open the source repository on GitHub and follow its README. clowder-ai is a agent — MCP Agents Market links you directly to the official repo.
Is clowder-ai free?+
clowder-ai is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.