gpu-cluster-managerTurn your spare GPUs into one inference endpoint. Auto-discovers machines on your network, routes requests to the best available device, learns when your mac...
Install via ClawdBot CLI:
clawdbot install twinsgeeks/gpu-cluster-managerGrade Fair — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Sends data to undocumented external endpoint (potential exfiltration)
POST → http://localhost:11435/dashboard/api/settingsCalls external URL not in known-safe list
https://github.com/geeks-accelerator/ollama-herdAudited Apr 17, 2026 · audit v1.0
Generated May 6, 2026
Combine spare GPUs from multiple machines (Mac Studio, MacBook, Linux box) into a single inference endpoint for running local LLMs. Automatically discovers machines via mDNS and routes requests to the best available device, ideal for self-hosted AI experimentation.
Leverage existing Apple Silicon and Linux hardware to create a GPU cluster for AI development without cloud costs. Helps startups and individual developers run models like Llama 3.3 locally with zero-config setup, reducing dependency on expensive cloud GPU rentals.
Pauses inference during video meetings (camera/mic detection on macOS) to free resources for productivity. Useful for remote workers or small teams who multitask between AI tasks and communication tools.
Deploy a distributed AI inference endpoint across multiple machines in a lab or office for edge computing. Routes requests based on capacity and latency, suitable for applications requiring low-latency local processing.
Set up a multi-machine GPU cluster for teaching AI/ML concepts in a classroom or workshop. Students can run models simultaneously without needing individual powerful hardware, with auto-discovery and simple commands.
Offer the basic cluster manager for free with limited node count or features. Charge for a Pro version with advanced analytics (per-app tracking, trends dashboard), priority support, and unlimited nodes.
Provide paid consulting services for enterprises to set up and optimize the cluster for their specific workflows, including integration with existing AI pipelines (LangChain, CrewAI) and custom model recommendations.
Partner with hardware vendors (e.g., Apple, NVIDIA) to bundle the cluster manager as a value-add for their devices. Generate revenue through licensing deals or referral fees per hardware sale.
💬 Integration Tip
For seamless integration, run `pip install ollama-herd` and start the router with `herd`, then add nodes with `herd-node`. Point any OpenAI-compatible client to `http://localhost:11435` to start using your cluster instantly.
Scored May 6, 2026
Search and manage Confluence pages and spaces using confluence-cli. Read documentation, create pages, and navigate spaces.
Manage torrents with qBittorrent. Use when the user asks to "list torrents", "add torrent", "pause torrent", "resume torrent", "delete torrent", "check download status", "torrent speed", "qBittorrent stats", or mentions qBittorrent/qbit torrent management.
Manage Notes, Tasks, Calendar, Files, and Contacts in your Nextcloud instance via CalDAV, WebDAV, and Notes API. Use for creating notes, managing todos and calendar events, uploading/downloading files, and managing contacts.
Search indexers and manage Prowlarr. Use when the user asks to "search for a torrent", "search indexers", "find a release", "check indexer status", "list indexers", "prowlarr search", "sync indexers", or mentions Prowlarr/indexer management.
Backup Clawdbot workspace and config to GitHub with git-crypt encryption. Use for daily automated backups or manual backup/restore operations.
Interact with Uptime Kuma monitoring server. Use for checking monitor status, adding/removing monitors, pausing/resuming checks, viewing heartbeat history. Triggers on mentions of Uptime Kuma, server monitoring, uptime checks, or service health monitoring.