Automation

You can plug a local LLM into your Gitea Actions, or any CI system, to review pull requests on its own. The pipeline pulls the diff, feeds it to a model running on Ollama , and posts structured feedback as PR comments. No code ever leaves your network. The setup needs three parts: a self-hosted runner with GPU access, a review prompt template, and a short Python wrapper.

Why Local LLM Code Reviews Make Sense

Static analysis tools like ESLint , Ruff , and Semgrep are great at catching syntax errors, style slips, and known vulnerability patterns. What they miss are logic bugs, unclear variable names, missing edge cases, and design concerns. An LLM fills that gap because it reads code in context. It can tell you that a function does the wrong thing, not just that it’s formatted wrong.

You can build a bot that labels issues, enforces PR naming, posts review comments, and triggers workflows. Write a FastAPI app that takes webhooks from GitHub or Gitea , checks the signature, and calls back to the right API. The same handler works for both forges. Header names and payload shape differ a bit, so one codebase can serve both.

How Repository Webhooks Work on GitHub and Gitea

Both GitHub and Gitea let you set up webhooks at the repo, org, or (for Gitea) system level. When an event fires (someone opens an issue, pushes a commit, opens a PR) the forge sends an HTTP POST to a URL you control. The body is JSON and describes what happened.

Running Home Assistant OS (HAOS) inside a Proxmox VE virtual machine gives you the full, officially supported installation - add-ons, Supervisor, automatic updates - while sharing hardware with other VMs and containers. On a modest Intel N305 mini PC, you can run HAOS alongside Plex, Vaultwarden, Nextcloud, and a WireGuard VPN with room to spare. The entire setup takes under 30 minutes. Download the HAOS QCOW2 image, create a VM in Proxmox, import the disk, boot, and you are up and running.

Home Assistant Blueprints are reusable automation templates. They split the logic from the per-device bits. You define a pattern once, say a motion light with a timeout, then spin it up for every room by filling in a form. No YAML to copy. No ten near-twin automations to babysit. In Home Assistant 2026.4, blueprints span three domains: automation, script, and template. They ship dozens of selector types for clean input forms and tidy collapsible sections for bigger setups. They’re the fastest way to keep smart home behavior the same across many devices.

Yes, Home Assistant ships a built-in conditional card. It shows or hides any dashboard card based on live state: entity value, time of day, who is home, screen size, and more. Add template sensors and a few HACS cards, and you can build dashboards where morning shows weather and coffee buttons, evening shows media and light scenes, and an empty house shows cameras and alarm controls. Cards pop in and out without leaving blank gaps, all through the stock Lovelace frontend. No custom code needed.

You can use HDMI-CEC commands through Home Assistant ’s HDMI-CEC integration - or a CEC-capable device like a Raspberry Pi running cec-client - to control TV power, input switching, and volume from automations and dashboards. Instead of juggling three or four remotes, you wire up a “Movie Mode” automation that dims the lights , powers on the TV, switches to the correct HDMI input, and sets volume to a comfortable level. One tap. Done.