A command-line framework for testing Android apps and devices with SoloPi, including on-device or cloud AI decision models. It manages devices, test cases, recorded interactions, replays, performance history, and evidence.

Checking devices and app capabilities, creating and validating test cases, recording and replaying interactions, running batches or stress tests, managing models, and exporting test evidence.

It turns requirements and acceptance criteria into repeatable test runs with three possible outcomes: pass, fail, or needing review. It also supports unattended execution across devices and performance monitoring.

solopi-ai — Skill for Codex by alipay/SoloPi

About the project

SoloPi is an open-source Android automation and testing system that records, replays, verifies, and diagnoses interactions on real devices. It is for mobile developers and testers who need repeatable UI tests, performance measurements, controlled execution, and evidence-based results in development or CI workflows. Its catalogue skill is the agent-facing entry point for submitting controlled SoloPi test tasks.

alipay/SoloPi · 6,348 stars · on GitHub

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 alipay/SoloPi --skill solopi-skill

Clone the repo

git clone --depth 1 https://github.com/alipay/SoloPi

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/alipay/solopi/solopi-skill/github.svg)](https://agentmods.dev/skills/alipay/solopi/solopi-skill)

Your own site

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

Per session 82 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 6,794 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. Third-party audits

NVIDIA SkillSpector pass 7 Sept 2026

How audits are shown

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	$0.00082	$0.06794
Opus 5.5	$0.00033	$0.02718
Sonnet 5.5	$0.00016	$0.01359
Haiku 4.5	$0.00008	$0.00679

Measured 8d ago against content hash 6d10c5a0b6ba, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

solopi-ai 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 8d 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.

solopi-skill/SKILL.md · 237 lines

How it starts

The opening of the file, as written. The whole thing — 237 lines — stays where its author put it; the contents beside it link to each section on GitHub.

SoloPi AI 控制框架

将本文件所在目录作为 Skill 根目录。只使用 Skill 的薄入口：

scripts/solopi-ai --pretty <命令>

执行时把 scripts/solopi-ai 解析为绝对路径，并保持调用者当前工作目录不变。Harness 实现与内置模板随本 Skill 放在 scripts/harness/，入口默认只调用这一份实现，不依赖仓库外的同级目录或已安装的 Python 包。调用方显式配置 SOLOPI_HARNESS_CLI 时可指定另一准确 CLI；普通使用无需配置。不得自行拼接 HTTP、Scheme 或 Shell 命令代替类型化 CLI。

安装时完整复制本 Skill 目录，保留脚本执行权限；运行需要 Python 3.9+ 和设备操作所需的 ADB。Laya 权重、虚拟环境、APK 和真机证据不随 Skill 打包。独立安装与本地验证见内置工作流。

全局参数必须放在子命令前。多台设备在线时，每次调用都必须添加 --serial <adb-serial>。

快速使用

用户要“连接手机跑几个案例”“运行支付宝模板”“查看带截图的结果”时，优先读取内置工作流，使用以下入口：

scripts/solopi-ai --pretty workflow-list
scripts/solopi-ai --serial <adb-serial> --pretty workflow-run --template alipay-more-services
scripts/solopi-ai --pretty workflow-report --artifacts <上一步返回的artifacts>

默认规则模式不加载模型。只有用户明确选择 Laya 实验时，才传 --mode laya --model <本地MLX权重目录>；模型运行在电脑上。不要为这些模板安装 companion APK，也不要把模型输出当成验收结论。 workflow-run 自动完成预检、计划编译与校验、执行及截图报告；前置页面不满足时保留 not_tested，不自动重跑。完成后返回 Markdown 报告路径，并说明三态结果、清理结果和截图是否齐全。

能力路由

需要判断某项 App 能力是直接 CLI、用例间接、用户交互还是暂不可安全自动化时，先读取能力矩阵。需要证明页面、设置和动作是否已纳入覆盖时，再读取 App 功能覆盖清单。

用户意图	使用命令	必须读取
选择内置模板、一条命令执行、查看整屏截图报告	`workflow-list`、`workflow-run`、`workflow-report`	内置工作流
检查设备、恢复内部 ADB、查询版本和协议	`doctor`、`adb-connect`、`capabilities`、`app-info`、`app-status`	故障处理
发现被测应用与动作	`apps`、`actions`	CLI 参考、动作目录
查询或修改配置	`config-list`、`config-get`、`config-set`	CLI 参考、协议
启动、观察或停止交互录制	`record-start`、`record-status`、`record-stop`	CLI 参考、故障处理
查询、导出或删除用例	`cases`、`case-get`、`case-delete`	CLI 参考
创建、校验或导入用例	`case-template`、`case-validate`、`case-import`	用例编写、动作目录
查看或编辑导出用例的步骤	`case-step-list`、`case-step-add`、`case-step-update`、`case-step-delete`、`case-step-move`、`case-step-copy`	用例编写、动作目录
获取当前页面和真实选择器	`inspect`	用例编写
单次同步或异步回放	`run`、`status`、`result`、`cancel`	协议、CLI 参考
动态 Agent 观察、类型化动作、暂停与清理	`agent-session-start`、`agent-observe`、`agent-act`、`agent-status`、`agent-timeline`、`agent-pause/resume/end/cancel`	协议、CLI 参考
从需求/AC 编译并执行可复现验证	`verify-normalize`、`verify-compile`、`verify-validate`、`verify-run`	验证工程、CLI 参考
持久设备池、任务队列、矩阵调度和 CI/API 托管执行	`managed-*`	托管执行与设备池、协议
签名端侧模型安装、激活、回退、推理、基准与发布门禁	`model-*`、`verify-run --decision-provider`	端侧 Agent 模型部署、验证工程
重复或批量回放	`run-repeat`、`run-batch`	CLI 参考、协议
回放历史的列表、详情与删除	`replay-history-list`、`replay-history-get`、`replay-history-delete`	CLI 参考、协议
性能指标、实时监控和采集会话	`perf-list`、`perf-current`、`perf-display-start/status/stop`、`perf-start/status/stop`	性能测试
本地性能 CSV 描述性统计	`perf-analyze`	性能测试
安卓应用冷启动或暖启动耗时	`startup-time`	性能测试、CLI 参考
性能历史的列表、详情与删除	`perf-history-list`、`perf-history-get`、`perf-history-delete`	性能测试、协议
CPU 或内存压力测试	`stress-start`、`stress-status`、`stress-stop`	性能测试、故障处理
独立录屏会话与 MP4 证据	`screen-record-start`、`screen-record-status`、`screen-record-stop`	CLI 参考、故障处理
相机扫码会话与只读内容	`scan-start`、`scan-status`、`scan-cancel`	CLI 参考、协议、故障处理
录屏视频差分视觉响应耗时	`video-analysis-start`、`video-analysis-status`	性能测试、CLI 参考
插件列表、本地安装与精确移除	`plugin-list`、`plugin-install`、`plugin-remove`	CLI 参考、故障处理
独立采集截图或日志	`screenshot`、`logs`	CLI 参考

Read the full file on GitHub · 237 lines

Files

What ships with it

60 files 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.

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.

8d ago Changed · +19 lines · -45 tokens per session 6d10c5a0b6ba
1mo ago First seen · 218 lines · 127 tokens per session scan A 4061512e6532

Subscribe to this mod's changes

solopi-ai is a skill published in the GitHub repository alipay/SoloPi (6,348 stars, last pushed yesterday), licensed Apache-2.0. It adds 82 tokens to every session and 6,794 once invoked, about $0.0003 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 2026-08-30.

E2E testing Mobile Testing adb android automated-testing automation mobile-testing ui-testing

mobile-tester

You are the Mobile Tester Specialist. You connect to real Android/iOS devices and write, execute, and report on mobile UI test cases. You master THREE testing modalities.

buiphucminhtam/forgewright · 81 tokens

android-ui-journey-testing

XML-specified Android UI journey testing, interactive step execution, assertion verification, and JSON outcome reporting.

sickn33/agentic-awesome-skills · 27 tokens

Espresso Android Testing

Android UI testing with Espresso framework including view matchers, actions, assertions, idling resources, and RecyclerView testing patterns.

PramodDutta/qaskills · 30 tokens

android-ui-test

Demonstration skill showing how to use uiautomator2-mcp-server for AI-driven Android UI testing. Also serves as the project's own automated UI test suite for continuous validation.

tanbro/uiautomator2-mcp-server · 41 tokens

android-qa

Test an Android app on a real phone or an emulator using adb. Use when the user wants to install an APK they just compiled, walk through the app's screens, check the on-screen text, reproduce a bug, capture screenshots, read crash logs, or set up an Android testing environment from scratch. Covers both a physical…

willbytee-sudo/android-qa-kit · 80 tokens

playwright-for-android

End-to-end Android debug loop on an emulator. Builds the APK with a Gradle task, boots a named AVD, installs and launches the target package, then drives a single feature via adb-injected UI actions while capturing a unified logcat-and-actions timeline plus a developer-facing test report. Use when the user wants to…

fclab-skku/playwright_for_android · 97 tokens