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.
npx skills add nolte/claude-home-assistant --skill ha-integration-scaffoldgit clone --depth 1 https://github.com/nolte/claude-home-assistantWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/nolte/claude-home-assistant/ha-integration-scaffold)<a href="https://agentmods.dev/skills/nolte/claude-home-assistant/ha-integration-scaffold"><img src="https://agentmods.dev/badge/skills/nolte/claude-home-assistant/ha-integration-scaffold/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/nolte/claude-home-assistant/ha-integration-scaffold"><img src="https://agentmods.dev/badge/skills/nolte/claude-home-assistant/ha-integration-scaffold.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 5.1 | $0.00153 | $0.03135 |
| Opus 5 | $0.00077 | $0.01568 |
| Sonnet 5 | $0.00031 | $0.00627 |
| Haiku 4.5 | $0.00015 | $0.00314 |
Grade A, and why
ha-integration-scaffold 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 11d 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.
How it starts
The opening of the file, as written. The whole thing — 178 lines — stays where its author put it; the contents beside it link to each section on GitHub.
HA Integration Scaffold
Spec: spec/claude/ha-integration-scaffold/en.md (EN canonical) / spec/claude/ha-integration-scaffold/de.md (DE translation).
Why this is a skill, not an agent
- One step in a larger human-driven workflow (decisive): the greenfield scaffold is the opening move of integration development; domain, platforms, discovery, and harness choices are elicited and approved mid-flow.
- Context access: the scaffold lands in the repository the conversation is scoped to and its decisions feed directly into the follow-on
ha-*skills. - Counter-dimension considered: a complete skeleton is a large self-contained unit (agent bias), but every structural decision inside it is operator-facing; a fire-and-forget agent would guess where the spec demands a confirmed answer.
When this skill activates
Use this skill when the user wants to:
- bootstrap a brand-new HA Custom Integration in an empty consumer repo
- get a runnable, lint-clean, test-passing skeleton without having to read every
spec/ha/*spec first - start an integration that targets HACS distribution
When NOT to activate
- editing an existing
custom_components/<domain>/→ route to the owning augment skill (ha-config-flow-augment,ha-coordinator-add,ha-entity-platform-add, …) or letha-integration-solutionplan the combination - scaffolding a Lovelace card → separate skill
ha-lovelace-card-scaffold - scaffolding an ESPHome device config (YAML) →
ha-esphome-config-scaffold; an ESPHome custom component (C++/Python) stays out of scope - scaffolding a blueprint / automation → separate skill
ha-blueprint-scaffold - migrating a YAML-configured integration to config flow → separate skill (planned, only on demand)
Hard rules
- Never overwrite an existing
custom_components/<domain>/. If the directory exists, abort with the path quoted. Collision is a user-disambiguation problem, not a generator problem. - Never use
hass.data[DOMAIN]. Every setup artefact lives in typedentry.runtime_data(seespec/ha/runtime-data-pattern/en.md). The generated__init__.pyis the single source of this contract. - Never set
_attr_name = "<string>"on a generated entity. All entity names live instrings.jsonunderentity.<platform>.<translation_key>.name(seespec/ha/translations/en.md). Hard-coded names break translation and stableentity_idslugs. - Never set
self.entity_id = "...". HA derives theentity_idfrom the system-language display name at first registration. The skeleton stays English-source so the slug remains language-independent (seespec/ha/entity-architecture/en.md). - Never bypass the API path whitelist. Generated
api.pycarries_API_PATH_REand a_with_auth(headers)helper; bearer tokens go on the wire only after the path has passed the whitelist (seespec/ha/security-hardening/en.md). - Never silently default. When the user did not specify
hacs/zeroconf/auth/platforms, fall back to documented defaults — but state every default in the output summary so the user sees what was assumed. - Name every artefact per
ha/naming-conventions. The integrationdomain,unique_id,translation_key, deviceidentifiers, service names, config-entry title, and all generated file paths follow the consolidated naming authority —snake_caseidentifiers, English display names (≤ 50 chars), no volatile data (IP, hostname, token, timestamp) in any ID (seespec/ha/naming-conventions/en.md). Rules 3–4 are the entity-name instances of this authority. - Verify HA internals against the official docs. Don't reproduce HA API signatures, lifecycle hooks, conventions, or schemas from memory — when uncertain, consult the official docs before generating or relying on it: Developer docs
developers.home-assistant, architecture/blueprint/YAML docshome-assistant.io(seespec/ha/upstream-docs-verification/en.md). - Emit
PARALLEL_UPDATESin every platform module. Each generated entity-platform module carries a module-levelPARALLEL_UPDATESconstant —0for coordinator-backed read-only platforms — so the integration satisfies the Silver quality-scale ruleparallel-updatesthatha-quality-scale-auditchecks. Verify the exact numeric convention against the HAparallel-updatesrule page.
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.
- 11d ago First seen · 178 lines · 153 tokens per session scan A 1eacac18680d
ha-integration-scaffold is a skill published in the GitHub repository nolte/claude-home-assistant (1 stars, last pushed 1mo ago), licensed MIT. It adds 153 tokens to every session and 3,135 once invoked, about $0.0008 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 2026-08-31.
Other skills, from other repositories
embedded-iot
Embedded systems firmware, microcontrollers (ESP32, STM32, Arduino, Raspberry Pi), RTOS (FreeRTOS, Zephyr), IoT protocols (MQTT, CoAP, BLE), bare-metal C/C++, and hardware peripheral interfaces (I2C, SPI, UART, GPIO). Use when developing firmware, working with microcontrollers, or building IoT devices.
spark-environment-setup
Set up a working ML training/inference environment on NVIDIA DGX Spark (GB10, aarch64, CUDA 13). Use when installing PyTorch/Unsloth/TRL/vLLM on DGX Spark, hitting libcudart or wheel-ABI errors on aarch64, or choosing between NGC containers and bare pip installs.
spark-training-gotchas
Preflight and diagnose the ten known failure modes for ML training on NVIDIA DGX Spark. Use when a training run on DGX Spark fails to start, OOMs below the 128GB limit, slows down mid-run, or before any multi-hour training job on GB10.
security-compliance
Guides security professionals in implementing defense-in-depth security architectures, achieving compliance with industry frameworks (SOC2, ISO27001, GDPR, HIPAA), conducting threat modeling and risk assessments, managing security operations and incident response, and embedding security throughout the SDLC.
stride-analysis-patterns
Apply STRIDE methodology to systematically identify threats. Use when analyzing system security, conducting threat modeling sessions, or creating security documentation.
review-loop
Run the adversarial verification loop — implement, then hand the change to a fresh checker that did not write it, fix what it finds, and re-dispatch until APPROVE. Use before claiming any behavioural change is done, and on requests like "review loop", "adversarial review", "independent review", "get this verified"…