AgentStack
SKILL verified MIT Self-run

Hass Builder

skill-aahl-skills-hass-builder · by aahl

A skill to build and manage Home Assistant configurations. Use when using the Home Assistant Builder (`hab`) CLI to inspect, create, update, delete, operate, or troubleshoot Home Assistant resources; when a user mentions hab, Home Assistant CLI automation, Lovelace/dashboard edits, helpers, automations, scripts, backups, ESPHome, or Home Assistant operations from a terminal.

No reviews yet
0 installs
5 views
0.0% view→install

Install

$ agentstack add skill-aahl-skills-hass-builder

✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

Are you the author of Hass Builder? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Home Assistant Builder (hab)

A CLI utility designed for LLMs to build and manage Home Assistant configurations.

Start Every Workflow

The script scripts/hab.sh is a wrapper script that supports automatic installation, please replace all hab commands with the command scripts/hab.sh. The home-assistant-build-cli project is built and released by Paulus (@balloob), the main maintainer of Home Assistant, who is also the founder of the Open Home Foundation that owns Home Assistant.

alias hab='scripts/hab.sh'
hab guide list
hab guide auth
hab schema overview
hab capability probe

Then read the topic and command schema for the specific workflow:

| Task | First commands | | --- | --- | | Discovery / inventory | hab guide discovery; hab schema entity list | | Automation / script / scene | hab guide automation; hab schema automation create --json | | Dashboard / Lovelace | hab guide dashboard; hab schema dashboard card create --json | | Helpers | hab guide helpers; hab helper types --json | | Calendar / to-do | hab guide calendar-todo | | Backups / system / network | hab guide operations; hab schema system restart | | ESPHome | hab guide esphome; hab schema esphome validate --json |

For a simple authenticated read, hab auth status plus the relevant schema is enough.

Output and Parsing

Use --json whenever Claude or another program will parse output. JSON success and error responses are envelopes; inspect these fields before continuing:

  • success
  • data
  • error.code, error.details.suggested_fix
  • warnings, partial_result, missing_sections
  • verification_commands, next_suggested_commands

ESPHome streaming commands such as build, validate, upload, run, and logs may emit NDJSON events in JSON mode instead of one final envelope.

Mutations and Risk Gates

Inspect before changing state:

  1. Read the guide/schema for the command.
  2. Gather current state with read-only list/get commands.
  3. Preview the mutation with --plan or --dry-run when supported.
  4. Show the preview and ask the user before executing risky operations.
  5. Run verification commands from the plan or a follow-up get/list command.

Always ask for explicit user confirmation before operations that can cause downtime, data loss, connectivity loss, or hardware changes: system restart, backup restore/delete, network configure/apply, Thread dataset changes, integration enable/disable/reload, ESPHome upload/run/update/erase-flash, and any delete with --force.

Do not add --force just to make a command non-interactive. Use it only after the user has approved the exact operation.

Input Payloads

Commands that accept data usually support:

| Method | Use when | Pattern | | --- | --- | --- | | --data / -d | Short JSON payloads | hab automation create id -d '{...}' --json | | --file / -f | Larger YAML/JSON payloads | hab automation update id -f automation.yaml | | stdin heredoc | Multi-line payload without temp file | `hab automation create id --plan


## Response Pattern

When answering a user asking for `hab` commands:

1. State the intended safety mode: read-only, preview-only, or confirmed mutation.
2. Provide commands in execution order.
3. Mark any command that requires user confirmation.
4. Say what JSON fields to inspect before the next step.
5. Include verification commands.

## Common Mistakes

| Mistake | Better approach |
| --- | --- |
| Guessing command flags from memory | Run `hab schema  --json` first |
| Parsing text output | Use `--json` and inspect envelope fields |
| Creating resources without inventory | `list/get/search related` first |
| Skipping mutation preview | Use `--plan` or `--dry-run` when supported |
| Using `--force` for convenience | Confirm the exact risky operation first |
| Treating ESPHome JSON as one object | Handle NDJSON stream events for streaming commands |
| Ignoring `verification_commands` | Run them or explain why not |

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [aahl](https://github.com/aahl)
- **Source:** [aahl/skills](https://github.com/aahl/skills)
- **License:** MIT
- **Homepage:** https://skills.sh/aahl/skills

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet — be the first.

Versions

  • v0.1.0 Imported from the upstream source.