My local LLM was just a chat box until Hermes Agent let it run scripts, files, and jobs for me
…Sign in to your XDA account I've been using Hermes Agent for a while now, and the more time I spend with it, the less I think about it as another…
Search
n8n, Dify, and Ollama might be the best self-hosted AI automation stack right now
…It focuses on agentic workflows, RAG pipelines, and deployment. You can expose what you build as APIs, which makes it easy for n8n to plug into. Ollama sits at the bottom as…
I let two local LLMs fight over how to optimize a Linux VM, and they destroyed it instead
…Qwen made the resolver nonsense and then claimed credit for fixing it. Gemma's summary of that run, written a few turns later, is great: "Agent A has effectively turned this machine…
Tailscale is the only home lab change I made this year that I actually noticed
…Each agent is also tracked by Tailscale's ACLs and metrics, and I can see which ones are using the most tokens. Related AI agents are a security nightmare for home labs…
Claude Cowork freed up 60GB on my laptop by finding files I completely forgot about
…However, keep in mind that this feature is currently in research preview, and Anthropic is still working on agent safety. The feature is also exclusive to Claude's paid plans for now…
Home Assistant finally has an AI tool worth the complexity, if you can get past the onboarding
…That excuse only works for so long The safety argument is valid, but it cannot serve as a permanent shield for poor onboarding. There is a difference between deliberate setup and scattered…
Claude's newest model is a step forward and two steps back, and it's infuriating
…Just in the past few days alone, the company launched Claude Design, routines in Claude Code, Claude for Word, Managed Agents, released Cowork generally and the list goes on. For many, the…
Claude Cowork handles my admin work so well that I stopped doing it myself
…can manipulate what an agent does on your behalf without you realising it. Anthropic’s own safety documentation flags web content and email as the primary surfaces for this. This is what…
I set up a Tailscale exit node at home, and it's more useful than just exposing services
…After all, it’s easy to configure even on CGNAT-afflicted setups and secure enough for anxious tinkerers like yours truly. However, the Tailscale ecosystem includes a bunch of quality-of-life…
I stopped using Claude for coding, but now I can't live without it for everything else
…I also feel meaningfully better using Claude for the personal, exploratory, or sensitive stuff because of Anthropic's stance on safety. Then there's Projects, which is one of the most underrated…