What is mobile-skill?

A tool for observing and controlling a real Android phone connected through ADB, a standard Android debugging connection.

Use it to open apps, tap, swipe, type, navigate, inspect installed apps, and handle installation or user-only prompts.

It gives the agent a defined way to inspect the phone, operate apps, and diagnose connection problems without relying on direct ADB commands.

mobile-skill — Skill for Codex by gpengzhi

Install

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.

Any agent

npx skills add gpengzhi/mobile-skill --skill skill

Clone the repo

git clone --depth 1 https://github.com/gpengzhi/mobile-skill

Made for: Codex.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/gpengzhi/mobile-skill/skill/github.svg)](https://agentmods.dev/skills/gpengzhi/mobile-skill/skill)

Your own site

<a href="https://agentmods.dev/skills/gpengzhi/mobile-skill/skill"><img src="https://agentmods.dev/badge/skills/gpengzhi/mobile-skill/skill/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/gpengzhi/mobile-skill/skill"><img src="https://agentmods.dev/badge/skills/gpengzhi/mobile-skill/skill.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 48 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 3,331 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00048	$0.03331
Opus 5.5	$0.00019	$0.01332
Sonnet 5.5	$0.00010	$0.00666
Haiku 4.5	$0.00005	$0.00333

Measured 2mo ago against content hash ecb316450ae8, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

mobile-skill 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.

skill/SKILL.md · 172 lines

How it starts

The opening of the file, as written. The whole thing — 172 lines — stays where its author put it; the contents beside it link to each section on GitHub.

mobile-skill

Use msk for every device observation and action. Run exactly one msk command per shell invocation. Do not call ADB directly or generate control scripts.

Task Boundaries

Define the user's success condition before acting. Complete only the requested bounded task, stop as soon as its observable success condition is met, and do not add post-success navigation or "just in case" verification. If a required step keeps failing without progress, stop retrying and use request-help or report the blocker.

Efficiency Defaults

Discover the exact package with apps list --user-visible when needed, then inspect that package's app registry before GUI navigation. If the target app is not installed, open the system app market, search for the app, and install it before continuing; after installation, rerun apps list --user-visible and verify the exact package. Pause for login, permission, payment, or other user-only prompts rather than bypassing them. Use a curated deep-link template whenever it directly reaches the requested destination; otherwise use GUI navigation. Do not reject a verified destination merely because it contains user-supplied parameters.
When two or more independent visible targets are stable in the latest observation, use one bounded sequence of at most five actions. Split actions only when an earlier action may navigate, open a dialog, change the layout, invalidate coordinates, or make the next action conditional. Always verify the resulting observation.
For side-effecting actions, a sequence is allowed only when the user has authorized all of them and the targets are independent and remain fixed in the same frame. Otherwise perform and verify each action separately.

Before Starting

Before the first Session, after environment changes, or when diagnosing failures, run the relevant check:

msk --json doctor --agent HARNESS_NAME

Replace HARNESS_NAME with your registered harness (e.g., codex, claude-code, cursor). Use msk --json doctor when no Agent-specific check applies. Treat vision=unverified as normal; static checks cannot verify image understanding. Use msk devices and msk session list only for troubleshooting.

Read the full file on GitHub · 172 lines

Files

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.

agents/openai.yaml 289 B

Changes

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 · 172 lines · 48 tokens per session scan A ecb316450ae8

Subscribe to this mod's changes

mobile-skill is a skill published in the GitHub repository gpengzhi/mobile-skill (5 stars, last pushed 2mo ago), licensed MIT. It adds 48 tokens to every session and 3,331 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 2026-08-31.

Mobile

mobile-skill next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
mobile-skillgpengzhi/mobile-skill	Skill	5	2mo ago	MIT	A	48	Codex
orca-emulator-androidstablyai/orca	Skill	117k	yesterday	MIT	A	102	Claude Code, Codex
glance-canonical-layoutsandroid/skills	Skill	10k	yesterday	Apache-2.0	A	56	Claude Code
android-tombstone-symbolicationdotnet/skills	Skill	5.8k	yesterday	MIT	A	176	Claude Code
dogfoodcallstack/agent-device	Skill	5.3k	yesterday	MIT	A	55	Claude Code
winapp-mauimicrosoft/winappCli	Skill	1.4k	yesterday	MIT	A	68	Claude Code
app_remixrome-os/rome	Skill	889	yesterday	MIT	A	50	Claude Code, Codex

mobile-skill

Getting it into your agent

Wrote this? Show the measurements

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

mobile-skill

Task Boundaries

Efficiency Defaults

Before Starting

What ships with it

What this file has done since we first saw it

mobile-skill compared with skills from other repositories