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-automation-solutiongit 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-automation-solution)<a href="https://agentmods.dev/skills/nolte/claude-home-assistant/ha-automation-solution"><img src="https://agentmods.dev/badge/skills/nolte/claude-home-assistant/ha-automation-solution/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-automation-solution"><img src="https://agentmods.dev/badge/skills/nolte/claude-home-assistant/ha-automation-solution.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 | $0.02253 |
| Opus 5.5 | $0.00080 | $0.00901 |
| Sonnet 5.5 | $0.00040 | $0.00451 |
| Haiku 4.5 | $0.00020 | $0.00225 |
Grade A, and why
ha-automation-solution 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 2mo 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 — 125 lines — stays where its author put it; the contents beside it link to each section on GitHub.
HA Automation Solution
Spec: spec/claude/ha-automation-solution/en.md (EN canonical) / spec/claude/ha-automation-solution/de.md (DE translation).
This skill is the front door to the ha-automation skill family. It does not generate any artifact itself — it decomposes the requirement, plans the combination, and dispatches the owning authoring skills, each of which owns its generation and spec conformance.
Why this is a skill, not an agent
- Plan-before-generate gate — the artifact plan must be presented and explicitly approved before any generation; that human-visible gate is core to the contract and an agent's fire-and-forget shape would lose it.
- Mid-flow interactivity — clarifying questions, plan confirmation, and the "this actually needs a custom integration" decision are per-run dialogues.
- Orchestrator that dispatches other skills — the skill-orchestrates-skill default (see
skill-vs-agent) keeps the entry point in skill form, likeha-blueprint-scaffolddispatchingha-blueprint-author. - Counter-dimension considered: the per-artifact generation could run as parallel agents, but the plan approval and the entity_id threading must stay visible in the user's context; skill wins.
When this skill activates
Use this skill when the user describes a result that likely needs more than one artifact, and should not have to know which authoring skill produces what.
When NOT to activate
- a single clear artifact (one automation, one helper, one sensor, one blueprint) → let the owning skill activate directly
- a Python custom integration (own device/cloud protocol, polling, config flow) →
ha-integration-scaffold - deploying/importing into a running HA instance → out of scope (generation only)
Hard rules
- Never generate inline. Every artifact is produced by its owning skill, resolved at runtime from the live
ha-automationfamily inventory (see Runtime skill resolution) rather than a frozen name list — the skill names in the decomposition heuristic are illustrative anchors. This skill plans and dispatches; it does not write artifacts. - Plan before generate. Always present the dependency-ordered artifact plan and wait for explicit approval before dispatching anything.
- One requirement, one run. No multi-requirement batches.
- Minimal artifacts. Decompose to the fewest artifacts that satisfy the requirement; never add a helper or sensor a single artifact already covers.
- Thread identities. Dispatch in dependency order and pass the
entity_ids/identifiers produced in earlier steps as inputs to dependent steps. Keep all names consistent perspec/ha/naming-conventions/en.md. - Stop on NEEDS-WORK. If a dispatched skill returns NEEDS-WORK, stop and report — do not build a dependent artifact on an unfinished predecessor.
- Recognize integration-shaped work. When the requirement needs a custom integration rather than YAML, say so in the plan and point at
ha-integration-scaffoldinstead of forcing it into helpers/templates. - Verify HA internals against the official docs (see
spec/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.
- 2mo ago First seen · 125 lines · 201 tokens per session scan A 5afb3482552c
ha-automation-solution 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 2,253 once invoked, about $0.0008 per session on Opus 5.5. 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
technical-troubleshooting
Provide setup, troubleshooting, and maintenance guidance. Use when the user reports a device that won't power on, connectivity issues, setup questions, overheating, or maintenance concerns.
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.
agent-engineering-expert
Build LLM agents that use tools safely: tool design, the agent loop, memory, MCP servers, multi-agent orchestration, sandboxing and prompt-injection defence. Use when the user mentions AI agents, tool use or function calling, MCP or Model Context Protocol, autonomous workflows, multi-agent systems, LangChain or…
agent-embedded-developer
Specialist subagent: Embedded systems, firmware, RTOS, microcontrollers (STM32, ESP32, Arduino), IoT, and bare-metal C/C++ specialist. Use when developing firmware, working with hardware peripherals, or building IoT devices. Trigger phrases: embedded, firmware, RTOS, microcontroller, Arduino, ESP32, STM32, IoT…
opus-rail
Session control for the FULL opus-rail system (hook-enforced variant). "/opus-rail disable" (or off) pauses every rails injection for this session; "/opus-rail enable" (or on) resumes; "/opus-rail status" reports hook telemetry and mode. Use when the user runs /opus-rail with one of those arguments. This is NOT the…
flow-nexus-swarm
Cloud-based AI swarm deployment and event-driven workflow automation with Flow Nexus platform.