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 kaeawc/auto-mobile --skill auto-mobile-usagegit clone --depth 1 https://github.com/kaeawc/auto-mobileWrote 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/kaeawc/auto-mobile/auto-mobile-usage)<a href="https://agentmods.dev/skills/kaeawc/auto-mobile/auto-mobile-usage"><img src="https://agentmods.dev/badge/skills/kaeawc/auto-mobile/auto-mobile-usage/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/kaeawc/auto-mobile/auto-mobile-usage"><img src="https://agentmods.dev/badge/skills/kaeawc/auto-mobile/auto-mobile-usage.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 | $0.00055 | $0.00946 |
| Opus 5.5 | $0.00022 | $0.00378 |
| Sonnet 5.5 | $0.00011 | $0.00189 |
| Haiku 4.5 | $0.00006 | $0.00095 |
Grade A, and why
auto-mobile-usage 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.
How it starts
The opening of the file, as written. The whole thing — 46 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Driving devices with AutoMobile
1. Get a session first
- Acquire a device with
getAndroidorgetApple. It finds or boots one and returns asessionUuid. - A connection that acquired a device is bound to it, so later calls route to that session without repeating it. A different connection, or a CLI call, passes
sessionUuidon every call. - A device that a session holds takes control calls only from that session (
device_owned_by_other_sessionotherwise). Naming someone else'ssessionUuidnever takes the device over. setActiveDeviceand acquisition are the only calls that change which session an untargeted call uses. With several sessions, passsessionUuidordeviceId(ambiguous_session_selector).session_ownership_lostandno_active_device_sessionare permanent for that session: acquire a new one.
2. Observe before you act
- Call
observebefore the first action and again after each action that can change the screen. Coordinates and Android synthetics2-element ids are valid only for the observation that returned them. - Act on what
observereturned:tapOntakes aselector(text,elementId,testTag),swipeOnscrolls the screen or a container (lookForscrolls until an element appears),tapAttakes coordinates,sendKeystypes into the focused field. Do not guess coordinates. skeletonlists only rows you can act on. Static text (a status label, a validation error, a result message) is incontext; read it there to verify a result.waitForwithtextortextAny, orfor: "appear"/"textEquals"with onlytext, matches text in either array.- Ask
observefor a screenshot only when the hierarchy is not enough; images are the largest part of a result.
3. Resources versus tools
- Tools act or run a computation. Resources (
automobile:URIs) are cheap read-only state:automobile:toolslists every tool and whether it is enabled,automobile:devices/bootedlists booted devices,automobile:observation/latestis the latest observation. - Read a resource instead of calling a tool when you only need state. A sessionless read never takes a held device.
- A tool that is gated off does not appear in
tools/list. Enable it withsetToolEnabledor theenableToolsargument ofgetAndroidandgetApple.
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 Changed · +1 lines bcae03c7d85e
- 11d ago First seen · 45 lines · 55 tokens per session scan A 82907be4a57e
auto-mobile-usage is a skill published in the GitHub repository kaeawc/auto-mobile (55 stars, last pushed today), licensed Apache-2.0. It adds 55 tokens to every session and 946 once invoked, about $0.0002 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 2026-10-28.
auto-mobile-usage compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| auto-mobile-usagekaeawc/auto-mobile | Skill | 55 | today | Apache-2.0 | A | 55 | Claude Code, Codex |
| orca-emulator-androidstablyai/orca | Skill | 119k | today | MIT | A | 102 | Claude Code, Codex |
| glance-canonical-layoutsandroid/skills | Skill | 10k | 2d ago | Apache-2.0 | A | 56 | Claude Code |
| android-tombstone-symbolicationdotnet/skills | Skill | 5.8k | today | MIT | A | 176 | Claude Code |
| dogfoodcallstack/agent-device | Skill | 5.3k | today | MIT | A | 55 | Claude Code |
| winapp-mauimicrosoft/winappCli | Skill | 1.4k | today | MIT | A | 68 | Claude Code |
| app_remixrome-os/rome | Skill | 894 | today | MIT | A | 50 | Claude Code, Codex |