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-entity-platform-addgit 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-entity-platform-add)<a href="https://agentmods.dev/skills/nolte/claude-home-assistant/ha-entity-platform-add"><img src="https://agentmods.dev/badge/skills/nolte/claude-home-assistant/ha-entity-platform-add/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-entity-platform-add"><img src="https://agentmods.dev/badge/skills/nolte/claude-home-assistant/ha-entity-platform-add.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 | $0.00201 | $0.02978 |
| Opus 5 | $0.00101 | $0.01489 |
| Sonnet 5 | $0.00040 | $0.00596 |
| Haiku 4.5 | $0.00020 | $0.00298 |
Grade A, and why
ha-entity-platform-add 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 12d 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 — 120 lines — stays where its author put it; the contents beside it link to each section on GitHub.
HA Entity Platform Add
Spec: spec/claude/ha-entity-platform-add/en.md (EN canonical) / spec/claude/ha-entity-platform-add/de.md (DE translation).
Why this is a skill, not an agent
- Human-visible scaffolding surface — the operator describes a device capability and reads back the platform module, the feature bitmask, the command methods, and the conformance report; a skill keeps this on the visible command surface, like the sibling augment skills (
ha-coordinator-add,ha-entity-description-map,ha-device-automation-add). - Mid-flow interactivity — the platform/domain decision and the family confirmation are per-run dialogues the operator approves before generation.
- Bounded, inline generation — one platform module plus its description and setup fit inline; no isolated agent context is needed.
- Counter-dimension considered: the draft→validate loop could be an agent, but the domain decision and the active-vs-declarative gate belong in the operator's working context; skill wins.
When this skill activates
Use this skill to scaffold one active platform entity — a command-driven domain (climate, cover, light, fan, lock, media_player, vacuum, valve, humidifier, water_heater, siren, lawn_mower, …) whose entity exposes async command methods — into an existing integration.
When NOT to activate
- datapoints authored declaratively as
EntityDescriptiontables — read-typesensor/binary_sensor/button, and the description-table form ofnumber/select/switch/calendar/todowith no hand-written command/set method →ha-entity-description-map - the coordinator itself →
ha-coordinator-add - greenfield integration scaffolding →
ha-integration-scaffold - device-automation triggers/conditions/actions →
ha-device-automation-add - deploying/importing into a running HA instance → out of scope
Hard rules
- One platform entity, one run. No multi-platform batches.
- Read the operationalized spec first. Read
spec/ha/entity-platform-types/en.mdto pick active-vs-declarative and the family, then read the matching family spec (spec/ha/entity-platforms-controls/en.md/spec/ha/entity-platforms-climate/en.md/spec/ha/entity-platforms-devices/en.md/spec/ha/entity-platforms-media/en.md/spec/ha/entity-platforms-voice/en.md/spec/ha/entity-platforms-inputs/en.md/spec/ha/entity-platforms-sensors/en.md) in full. Do not generate from memory. - Active platforms only. This skill scaffolds command-driven entities authored as a full entity class with hand-written async command/set methods. The boundary is the authoring form, not the domain: a read-type datapoint (
sensor/binary_sensor/button) or any entity expressed purely as anEntityDescriptiontable without a hand-written command/set method → point atha-entity-description-mapand abort. Active platforms whose command/set method this skill writes — including theinputsfamily (number/select/text/date/time/datetime) andcalendar/todo— stay in scope. - Name the domain, confirm the family. Require the operator to name the target domain and confirm the resolved family before generating.
- Base class from the family spec. Derive the entity from the documented platform base class (
ClimateEntity/CoverEntity/LightEntity/FanEntity/LockEntity/MediaPlayerEntity/StateVacuumEntity/ValveEntity/HumidifierEntity/WaterHeaterEntity/SirenEntity/LawnMowerEntity…). - Feature bitmask from the enum. Set
supported_featuresas a bitwise|combination of the platform-native*EntityFeatureenum — never a raw integer. - Flag ↔ method, one-to-one. Implement the documented async command method for every set flag (e.g.
CoverEntityFeature.OPEN↔async_open_cover,ClimateEntityFeature.TARGET_TEMPERATURE↔async_set_temperature,LockEntityFeature.OPEN↔async_open,FanEntityFeature.SET_SPEED↔async_set_percentage). Never set a flag "on spec" whose method is missing. - Required properties, built-in enums. Provide every property the domain marks as Required (
hvac_mode/hvac_modes,is_closed,color_mode/supported_color_modes,activity,alarm_state, …) and use only the built-in state/mode enums (only built-inHVACMode;VacuumActivity/LawnMowerActivity). Setdevice_classfrom the closed platform-native enum where a member exists, never a free string. - Wire the setup. Implement
async_setup_entry(hass, entry, async_add_entities)that builds the entities and registers them viaasync_add_entities, attaching to the coordinator /config_entry.runtime_data; if no coordinator exists, point atha-coordinator-add. - Availability (
entity-unavailable, Silver). The generated entity handles availability — subclassCoordinatorEntity(which derivesavailablefromcoordinator.last_update_success) or override theavailableproperty to returnFalsewhen the datapoint can't be read or controlled. An active entity that never reports unavailable silently misses the Silverentity-unavailablerule thatha-quality-scale-auditchecks. - Emit
PARALLEL_UPDATES(parallel-updates, Silver). Declare a module-levelPARALLEL_UPDATESconstant in<platform>.py(a coordinator-backed read path typically uses0; command platforms bound their concurrency). Verify the value against the HAparallel-updatesrule page rather than reproducing it from memory. - Name per
spec/ha/naming-conventions/en.md, do not duplicate the generic entity pattern (delegate tospec/ha/entity-architecture/en.md), and verify HA internals against the official docs (seespec/ha/upstream-docs-verification/en.md).
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.
- 12d ago First seen · 120 lines · 201 tokens per session scan A 2f0df87a95d6
ha-entity-platform-add is a skill published in the GitHub repository nolte/claude-home-assistant (1 stars, last pushed 1mo ago), licensed MIT. It adds 201 tokens to every session and 2,978 once invoked, about $0.0010 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
authentication-patterns
OAuth 2.0, JWT, SSO, MFA, NextAuth/Clerk/Supabase Auth implementation patterns.
email-systems
Transactional email (Resend, SendGrid, SES), templates (React Email, MJML), deliverability (SPF/DKIM/DMARC), and inboxing best practices. Use when building email infrastructure, designing templates, or troubleshooting deliverability.
event-driven-architecture
Kafka, RabbitMQ, SQS/SNS, event sourcing, CQRS, saga patterns, dead letter queues, and idempotency. Use when designing asynchronous systems, implementing message-driven workflows, or building event streaming pipelines.
graphql-expert
GraphQL API design and implementation. Use when building GraphQL APIs, designing schemas, implementing resolvers, or optimizing GraphQL performance.
api-design
REST and GraphQL API design best practices including OpenAPI specs. Use when designing APIs, documenting endpoints, or reviewing API architecture.
generic-fullstack-feature-developer
Guide feature development for full-stack applications with architecture focus. Covers Next.js App Router patterns, NestJS backend services, database models, data workflows, and seamless integration. Use when adding new features, refactoring existing code, or planning major changes.