• Clawdbot + Docker Model Runner enables self-hosted, privacy-first personal AI assistants. • Integrates with Telegram, WhatsApp, Discord, Signal for proactive digital coworker. • Docker Model Runner runs LLMs as OCI artifacts, exposing OpenAI-compatible API. • Keeps all data, memory, and files on local hardware, preventing third‑party training. • Isolated sandboxed inference protects host system and ensures data sovereignty. • Offers cost control and scalable deployment for personal intelligence.

Article Summaries:

  • Clawdbot, a self‑hosted AI assistant that plugs into messaging apps such as Telegram, WhatsApp, Discord and Signal, can be paired with Docker Model Runner (DMR) to create a fully private personal assistant. DMR runs large language models as OCI artifacts and exposes an OpenAI‑compatible API, allowing Clawdbot to perform tasks-web browsing, file management, email summarisation-without sending data to external clouds. The stack keeps all memory, message history and files on the user’s own hardware, isolates inference in sandboxed containers, and eliminates per‑token costs by running inference locally on the user’s GPU/CPU. Users configure Clawdbot to point to DMR’s local endpoint (e.g., http://localhost:12434/v1) via JSON settings, enabling a privacy‑first, cost‑controlled AI assistant.

Sources: