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 agentmods add instructions/willbytee-sudo/android-qa-kit/agents-mdgit clone --depth 1 https://github.com/willbytee-sudo/android-qa-kitWhat 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 | $0.02707 | $0.02707 |
| Opus 5 | $0.01354 | $0.01354 |
| Sonnet 5 | $0.00541 | $0.00541 |
| Haiku 4.5 | $0.00271 | $0.00271 |
Grade A, and why
android-qa-kit AGENTS.md 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 yesterday.
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 — 265 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — setting up and driving an Android device
[!IMPORTANT] Agent: if you are reading this from a URL, fetch the raw file instead.
https://raw.githubusercontent.com/willbytee-sudo/android-qa-kit/main/AGENTS.mdThe rendered
github.compage loads long files in chunks. Fetching that HTML gets you roughly the first half and an abrupt ending, and you will conclude that sections are missing when they are not.Self-check: this file ends with the line "If you couldn't verify one of them, say so explicitly instead of assuming." If your copy ends anywhere else, you got a truncated read — refetch from the raw URL above, or
git clonethe repository and read the file locally.
This repository installs nothing on its own. There are no installers, cloning downloads nothing, and it does not touch the system. It is a guide: you, the agent, run each step in the open, explaining what it does and why.
Everything runs through adb, Android's official tool. No MCP required.
Read this fully before running the first command. There are six documented traps that break things which look correct, and one of them blinds you with no error output.
1. How to work with the user
You are their assistant for setting this up, not a silent installer. That means:
- One step at a time. Run it, show the result, and only then move on.
- Ask before downloading or installing anything. Say what it is, where it comes from, and how big it is.
- Verify by executing. An installer saying it finished means nothing — check that the binary exists and responds.
- If you couldn't verify something, say so. Don't assume it worked.
2. Start by finding out what's already there
adb devices
If adb isn't on the PATH, look for it before concluding anything is missing:
$paths = @("$env:ANDROID_HOME", "$env:LOCALAPPDATA\Android\Sdk", "C:\Android\Sdk")
$paths | ForEach-Object { if (Test-Path "$_\platform-tools\adb.exe") { "$_\platform-tools\adb.exe" } }
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.
- yesterday First seen · 265 lines · 2,707 tokens per session scan A d612a0950666
android-qa-kit AGENTS.md is an instructions file published in the GitHub repository willbytee-sudo/android-qa-kit (3 stars, last pushed 1mo ago), licensed MIT. It adds 2,707 tokens to every session, about $0.0135 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 instructions, from other repositories
android-agent CLAUDE.md
Instructions for ghost-in-the-droid/android-agent, covering ghost in the droid — agent tool rules, mcp tool: opencamera, mcp tool: speaktext and general rule.
compose-kotlin-agent-skills AGENTS.md
Instructions for haidrrrry/compose-kotlin-agent-skills, covering agents.md — compose-kotlin-agent-skills, build & validation, when you add or remove a skill.md, regenerate the lock file, skill catalog and reference modules (deep dives).
uiautomator2-mcp-server AGENTS.md
Instructions for tanbro/uiautomator2-mcp-server, covering uiautomator2-mcp-server, project overview, project structure, development setup and install dependencies (uses uv lock file).
uiautomator2-mcp-server CLAUDE.md
Instructions for tanbro/uiautomator2-mcp-server, a project described as: A MCP (Model Context Protocol) server that provides tools for controlling and interacting with Android devices using uiautomator2.
ThumbAgent AGENTS.md
AGENTS.md instructions for LiuShiYi1027/ThumbAgent, covering thumbagent agent guide, 1. 开始工作前, 2. v1 边界, 3. 强制架构边界 and 4. contract-first.
NeoAgent AGENTS.md
AGENTS.md instructions for NeoLabs-Systems/NeoAgent, covering codex instructions and additional notes for codex.