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 larsboes/Axon --skill esphomegit clone --depth 1 https://github.com/larsboes/AxonWrote 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/larsboes/axon/esphome)<a href="https://agentmods.dev/skills/larsboes/axon/esphome"><img src="https://agentmods.dev/badge/skills/larsboes/axon/esphome/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/larsboes/axon/esphome"><img src="https://agentmods.dev/badge/skills/larsboes/axon/esphome.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 | $0.00148 | $0.01251 |
| Opus 5.5 | $0.00059 | $0.00500 |
| Sonnet 5.5 | $0.00030 | $0.00250 |
| Haiku 4.5 | $0.00015 | $0.00125 |
Grade A, and why
esphome 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 — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.
esphome
Build and OTA-flash ESPHome nodes through the containerized esphome CLI on the home-automation
host. espctl is a thin wrapper: node builds/flashes run via docker exec over SSH; the dashboard
is polled directly only for a quick version reachability check. Node YAMLs live on the host in the
directory bind-mounted to /config in the container.
Container assumption
This drives a Docker container named esphome (override with $ESPHOME_CONTAINER) — typically
a service in the host's docker-compose/compose.yaml, with the host's ESPHome config dir mounted
to /config. Every command shells docker exec <container> esphome … over SSH; there is no bare
local ESPHome install in the loop. The dashboard normally listens on :6052. For OTA to work, the
ESPHome service must bind the LAN IP (--address in the compose service), and exposure of :6052
should be restricted to the LAN.
Setup (once)
- On the host: run the
esphomecontainer (compose service namedesphome) with the config dir mounted to/config, and ensure key-based SSH access to the host (the tool usesBatchMode=yes— no password prompt). - In the overlay config: the connection facts live in the committed overlay at
<overlay>/config/home-assistant.vars(path via$AXON_HOME_ROOT, default the active deployment overlay):ESPHOME_SSH(an ssh target such asuser@host),ESPHOME_URL(the dashboard base URL, e.g.http://host:6052), andESPHOME_CONTAINER(the container name). All three are non-secret and committed there; each can be overridden per-run via the matching$ESPHOME_*env var (env wins over the config file). - No secret for this skill. It reaches the container over SSH with key auth — there is no password or token, so there is no Bitwarden item to provision. (If a secret is ever needed later, add a bw fetch then; today there is none.)
Run
The tool reads its connection facts from the overlay config — no unlock step:
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.
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 · 82 lines · 148 tokens per session scan A f479497911c5
esphome is a skill published in the GitHub repository larsboes/Axon (1 stars, last pushed 1mo ago), licensed MIT. It adds 148 tokens to every session and 1,251 once invoked, about $0.0006 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.
esphome compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| esphomelarsboes/Axon | Skill | 1 | 1mo ago | MIT | A | 148 | Claude Code |
| usecase-smart-home-brainCoWork-OS/CoWork-OS | Skill | 537 | today | MIT | A | 22 | Claude Code, Codex |
| openhuetaracodlabs/aiden | Skill | 875 | 18d ago | AGPL-3.0 | A | 15 | Claude Code, Codex |
| freecadEldanRing/qubicl | Skill | 13 | 7d ago | Apache-2.0 | A | 23 | Claude Code, Codex |
| ryu-local-modelamajorai/ryu | Skill | 44 | today | none | A | 75 | Claude Code, Codex |
| beryl-it-techStunspot/nova-the-optimal-ai-mind | Skill | 31 | 10d ago | none | A | 17 | Claude Code, Codex |
| it-work-reviewerStunspot/nova-the-optimal-ai-mind | Skill | 31 | 10d ago | none | A | 17 | Claude Code, Codex |