Discover and explore top open-source AI tools and projects—updated daily.
pocketpawSelf-hosted personal AI agent platform
New!
Top 63.7% on SourcePulse
Summary
PocketPaw is a self-hosted, open-source personal AI agent platform designed for rapid deployment (under 30 seconds) and user privacy. It offers a modular, cross-platform solution with a web dashboard and integrations for various communication channels, allowing users to run AI agents locally without cloud lock-in. The platform targets power users and developers seeking a customizable, secure, and private AI assistant.
How It Works
PocketPaw supports multiple agent backends, including the Anthropic Claude Agent SDK (default) and Open Interpreter (for Ollama, OpenAI, Anthropic). It features a local-first architecture, storing data in markdown files (~/.pocketpaw/memory/) with optional semantic memory via Mem0. A "Smart Model Router" can dynamically select LLM models based on task complexity, optimizing performance and cost. Security is a core tenet, with features like a Guardian AI for command review and an Injection Scanner.
Quick Start & Requirements
Installation is streamlined via a single bash command (curl -fsSL https://pocketpaw.xyz/install.sh | sh or pip install pocketpaw && pocketpaw). Alternative methods include pipx, uvx, and Docker for server deployments. It runs natively on macOS, Windows, and Linux. Optional dependencies include Docker, and specific integrations may require API keys (e.g., Anthropic, Google, Tavily, OpenAI) or OAuth credentials. Local LLM inference can be enabled via Ollama.
Highlighted Details
Maintenance & Community
The project is explicitly marked as "Beta" and under active development, with users advised to expect breaking changes. Community links include Twitter (@PocketPawAI), with Discord mentioned as "Coming Soon." Contributions via Pull Requests are welcomed.
Licensing & Compatibility
Licensed under the MIT license, which is permissive for commercial use and integration into closed-source projects.
Limitations & Caveats
As a beta project, users should anticipate potential instability and breaking changes between versions. Certain advanced features or integrations may require significant additional setup or external API keys.
1 day ago
Inactive