メニュー
One change lap for Home Assistant Homelab: scope, edit, validate, hand off with a suggested Conventional Commit. Use when implementing a config fix, package change, plan checkbox, or agent-context update.
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/Apocrathia/home-assistant-config --skill implement-changeコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか？SkillsMP が現在取得できるファイルをダウンロードできます。
Mine Home Assistant history for automation trigger candidates from a narrative or structured event brief. Use when finding a reliable trigger, correlating a morning/evening sequence, ranking sensor signals for a new automation, or running the pattern-analyst ↔ pattern-reviewer pair.
Interview the operator about a topic until shared understanding, walking each branch of the decision tree one at a time. Use when stress-testing an idea, design, or requirement; resolving open questions before acting; or when the user says alignment, /alignment, or wants to get aligned.
Explore packages/, layout, and .agents/ for architectural friction (coupling, shallow packages, missing boundaries) and file or recommend agent issues under docs/issues/. Use when improving config shape, before large refactors, or when find-work should surface design-debt candidates.
SOC 職業分類に基づく
SKILL.md を表示中
| name | implement-change |
| description | One change lap for Home Assistant Homelab: scope, edit, validate, hand off with a suggested Conventional Commit. Use when implementing a config fix, package change, plan checkbox, or agent-context update. |
| disable-model-invocation | true |
One logical change from request to operator handoff. Edit only in the main
checkout (worktrees.md). Do not commit, push,
or open PRs unless the operator explicitly asks
(operator-owned-git.md).
| Surface | Route |
|---|---|
YAML packages / configuration.yaml | edit here; validate via config-validate |
| Package layout | package-organize |
| Broken automations | automation-debug |
| Docs issue / plan checkbox | implement slice; update plan checkboxes; delete-on-ship via reconcile-docs |
| Fuzzy scope | alignment first, then stop for operator |
Load context as needed: .agents/context/{constraints,traps,nomenclature,output,questions,vertical-slices,development-loop}.md.
Secrets: never hardcode; use !secret. Automations: modern keys
(triggers / conditions / actions; item keys trigger: / action:).
YAML: 2-space indent.
Plan/issue templates: docs/plans/_template.md, docs/issues/_template.md.
Ignore upstream steps that require worktrees or agent commits.
- [ ] 1. Frame scope (one change, done conditions; link issue/plan if any)
- [ ] 2. Edit in place (packages/, esphome/, docs/, .agents/ as needed)
- [ ] 3. Validate (config-validate + HA-oriented checks)
- [ ] 4. Update plan checkboxes / reconcile-docs if acceptance met
- [ ] 5. Hand off — stop for operator
One sentence goal + done conditions. Prefer one vertical slice
(vertical-slices.md). If two unrelated
outcomes, split. Protected paths need confirmation
(protected-paths.md).
Make the minimal change. Prefer existing package patterns under packages/.
Follow config-validate. For automations, also
check entity refs and modern key usage.
If working from a plan: tick completed checkboxes. If acceptance on the linked
issue is fully met, run reconcile-docs
delete-on-ship in this change.
## Implement change
**Goal:** <one sentence>
**Paths:** <files>
**Issue/plan:** <paths or none>
**Result:** ready for operator | blocked
### Validation
- <checks run + outcome>
### Suggested commit
`<type>(<scope>): <description>`
### Operator next
- Apply / reload HA as needed, validate live, then commit