A configuration guide for Home Assistant Assist, the voice-control system for smart homes. It covers the workflow from hearing speech to understanding commands and speaking a reply.

Use it to set up voice pipelines, custom spoken commands, wake words, and local speech-to-text or text-to-speech with tools such as Whisper and Piper. It also helps configure voice satellites, which are remote devices that listen for commands.

It helps connect speech recognition, command handling, wake words, and spoken responses into one voice workflow. It also shows how to tailor commands to the devices and rooms in a home.

ha-voice — Skill by nodnarbnitram

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods

npx agentmods add skills/nodnarbnitram/claude-code-extensions/ha-voice

Any agent

npx skills add nodnarbnitram/claude-code-extensions --skill ha-voice

Clone the repo

git clone --depth 1 https://github.com/nodnarbnitram/claude-code-extensions

Made for: Claude Code, Codex.

Per session 73 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 3,117 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5	$0.00073	$0.03117
Opus 5	$0.00036	$0.01558
Sonnet 5	$0.00015	$0.00623
Haiku 4.5	$0.00007	$0.00312

Measured 2d ago against content hash 48319b9d134d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

ha-voice scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/ha-voice/SKILL.md · 524 lines

How it starts

The opening of the file, as written. The whole thing — 524 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Home Assistant Voice Control

Configure Assist pipelines, custom intents, wake words, speech processing, and voice satellites for local voice control.

Quick Start

Setting up Assist Pipeline

# configuration.yaml
assist_pipeline:
  pipelines:
    - language: en
      name: Default Pipeline
      stt_engine: faster_whisper  # Local STT
      tts_engine: tts.piper       # Local TTS
      conversation_engine: conversation.home_assistant
      wake_word_entity: binary_sensor.wake_word

Why this matters: The pipeline connects all components (STT, TTS, intent, conversation agent) into a single voice workflow.

Creating Custom Intents

# custom_sentences/en/turn_on.yaml
language: en
version: 1

intents:
  TurnOn:
    data:
      - sentences:
          - "turn on the [area] {name:list}"
          - "turn on {name:list}"
          - "[please] activate {name:list}"
        lists:
          name: light_names

lists:
  light_names:
    - "bedroom light"
    - "kitchen overhead"
    - "living room"

Why this matters: Custom sentence patterns enable natural voice commands tailored to your home.

Configuring Text-to-Speech (TTS)

Local Option - Piper:

tts:
  - platform: piper
    language: en_GB
    voice: jenny_dioco

Cloud Option - OpenAI:

tts:
  - platform: tts
    service: google_translate_say  # or openai, google cloud
    language: en

Why this matters: TTS gives voice feedback; local options require no cloud dependencies.

Configuring Speech-to-Text (STT)

Local Option - Faster Whisper:

stt:
  - platform: faster_whisper
    language: en

Cloud Option - Google Cloud Speech:

stt:
  - platform: google_cloud
    language_code: en-US

Why this matters: STT converts spoken commands to text; local options provide privacy.

Critical Rules

✅ Always Do

✅ Use custom_sentences/ directory for intent configuration (not deprecated sentences/)
✅ Test intents with Assist Developer Tools before deploying
✅ Use {name:list} syntax for slot-based entity matching
✅ Configure both STT and TTS engines in the same pipeline
✅ Verify wake word entity exists before enabling in pipeline
✅ Document intent lists so voice commands are discoverable

Read the full file on GitHub · 524 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

README.md 5.9 KB

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

2d ago First seen · 524 lines · 73 tokens per session scan A 48319b9d134d

Subscribe to this mod's changes

ha-voice is a skill published in the GitHub repository nodnarbnitram/claude-code-extensions (16 stars, last pushed 4mo ago), licensed MIT. It adds 73 tokens to every session and 3,117 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Hardware and IoT

add-macos-statusbar

Add a macOS menu bar status indicator for NanoClaw. Shows a bolt icon with a green/red dot indicating whether NanoClaw is running, with Start, Stop, and Restart controls. macOS only.

nanocoai/nanoclaw · 49 tokens

cli-anything-3mf

3MF mesh geometry editor — detect and resize cylindrical holes, repair meshes, compare 3D printing files. Works with BambuStudio and PrusaSlicer 3MF files.

HKUDS/CLI-Anything · 44 tokens

platform-port

Guide porting FastLED to new MCU platforms, including int.h types, clockless drivers, SPI implementations, and platform detection. Use when adding support for a new microcontroller family or board.

FastLED/FastLED · 42 tokens

doca-compress

Use this skill for hands-on DOCA Compress programming on a BlueField DPU, ConnectX NIC, or host with DOCA — enabling compress-deflate, decompress-deflate, decompress-lz4-stream, or decompress-lz4-block tasks on a docacompress context (the hardware supports DEFLATE both directions plus LZ4 decompress; LZ4 encode is NOT…

NVIDIA/skills · 242 tokens

pcbway

PCBWay PCB fabrication and assembly — turnkey/consigned assembly, design rules, ordering workflow. Alternative to JLCPCB for manufacturing. Use with KiCad. Use this skill when the user mentions PCBWay, needs turnkey assembly (PCBWay sources parts by MPN), has parts not available on LCSC, needs assembled boards with…

aklofas/kicad-happy · 119 tokens

anta-validation

Validate Arista EOS network state with structured pass/fail verdicts using ANTA. Use for "is this switch healthy", "did my change break anything", "verify BGP/interfaces/hardware are correct", "run a health check on this device". Read-only. A test for a feature the device does not run reports notapplicable — never a…

automateyournetwork/netclaw · 84 tokens