aurora-smart-home
AGENTS.md109 starsChanged 10 months agoMITView on GitHub
Aurora Smart Home is a reference library for building and troubleshooting Home Assistant, ESPHome, Node-RED, and related smart-home projects. It ships as a Claude Code plugin, but the reference content underneath is plain Markdown, readable and usable from Codex too. There is no installer step. When a request touches one of the domains below, open the matching SKILL.md first: it is the entry point for that domain and links out to the detailed reference files (board/component data, official HA/ESPHome doc…
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 31,774 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository7 other files it gives its agents.
- API CatalogSkill
- auroraSkill
- ESPHomeSkill
- HA Dashboard DesignSkill
- HA Integration DevSkill
- Home Assistant YAMLSkill
- Node-REDSkill
The file
We split this AGENTS.md into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 29
Read it when a request could plausibly belong to more than one domain above.
≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
AGENTS.md
Aurora Smart Home is a reference library for building and troubleshooting Home Assistant, ESPHome, Node-RED, and related smart-home projects. It ships as a Claude Code plugin, but the reference content underneath is plain Markdown, readable and usable from Codex too.
How to use this repo from Codex
There is no installer step. When a request touches one of the domains below,
open the matching SKILL.md first: it is the entry point for that domain
and links out to the detailed reference files (board/component data,
official HA/ESPHome doc snapshots, recipes, troubleshooting guides). Treat
its instructions as authoritative for that domain, the same way you would
follow any other file this repo tells you to read.
| Domain | Read first |
| Not sure which domain, or it spans more than one | aurora/SKILL.md |
| ESPHome firmware (ESP32, ESP8266, Shelly, boards, sensors, BLE/Matter/Thread, audio, power) | esphome/SKILL.md |
| Home Assistant YAML (automations, scripts, dashboards, templates, blueprints) | home-assistant/SKILL.md |
| Custom Python integrations (HACS, config_flow, coordinators) | ha-integration-dev/SKILL.md |
| Node-RED visual flows | node-red/SKILL.md |
| External API integration (Tibber, SMHI, SL, Spotify, etc.) | api-catalog/SKILL.md |
| Dashboard visual styling (Mushroom, card-mod, themes) | ha-dashboard-design/SKILL.md |
aurora/SKILL.md also documents the routing logic in full (which specialist
persona owns which keywords, model recommendations, delivery conventions).
Read it when a request could plausibly belong to more than one domain above.
What does not carry over from Claude Code
This repo also ships Claude Code-specific plugin plumbing:
.claude-plugin/marketplace.json, hooks, and the /aurora:aurora slash
command. None of that applies to Codex; it is packaging for the Claude Code
plugin marketplace, not part of the reference content. Codex should read the
SKILL.md files directly as documentation, as described above.
Repo-wide conventions
Everything committed to this repo (docs, code, commit messages) is written
in English, even though maintainers may discuss it in other languages. See
CLAUDE.md for the full contributor rules (copyright/attribution handling
for third-party doc content, commit message style, etc.). They apply
regardless of which agent is doing the writing.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.