dogfood by callstackincubator/agent-device
npx skills add https://github.com/callstackincubator/agent-device --skill dogfood系统性地探索移动应用,发现问题,并为每个发现生成包含完整复现证据的报告。
仅 目标应用 是必需的。其他所有内容都有合理的默认值。
| 参数 | 默认值 | 示例覆盖 |
|---|---|---|
| 目标应用 | (必填) | Settings、com.example.app、深度链接 URL |
| 平台 | 从用户上下文推断;否则询问 (ios 或 android) | --platform ios |
广告位招租
在这里展示您的产品或服务
触达数万 AI 开发者,精准高效
| 会话名称 | 应用/平台的简化名称(例如 settings-ios) | --session my-session |
| 输出目录 | ./dogfood-output/ | 输出目录: /tmp/mobile-qa |
| 范围 | 完整应用 | 专注于引导流程和个人资料 |
| 认证 | 无 | 登录到 user@example.com |
如果用户提供了足够的上下文来开始,则立即使用默认值开始。仅当缺少必需细节(例如平台或凭据)时,才询问后续问题。
优先使用直接的 agent-device 二进制文件(如果可用)。
1. 初始化 设置会话、输出目录、报告文件
2. 启动/认证 打开应用并在需要时登录
3. 定位 捕获初始快照并映射导航
4. 探索 系统性地测试流程和状态
5. 记录 为每个问题记录可复现的证据
6. 收尾 汇总摘要,关闭会话
mkdir -p {OUTPUT_DIR}/screenshots {OUTPUT_DIR}/videos
cp {SKILL_DIR}/templates/dogfood-report-template.md {OUTPUT_DIR}/report.md
启动一个命名会话并打开目标应用:
agent-device --session {SESSION} open {TARGET_APP} --platform {PLATFORM}
agent-device --session {SESSION} snapshot -i
如果需要登录:
agent-device --session {SESSION} snapshot -i
agent-device --session {SESSION} fill @e1 "{EMAIL}"
agent-device --session {SESSION} fill @e2 "{PASSWORD}"
agent-device --session {SESSION} press @e3
agent-device --session {SESSION} wait 1000
agent-device --session {SESSION} snapshot -i
对于 OTP/邮件验证码:询问用户,等待输入,然后继续。
捕获初始证据和导航锚点:
agent-device --session {SESSION} screenshot {OUTPUT_DIR}/screenshots/initial.png
agent-device --session {SESSION} snapshot -i
在进行深度测试之前,映射顶层导航、标签页和关键工作流程。
阅读 references/issue-taxonomy.md 以进行严重性/类别校准。
策略:
diff snapshot -i 以避免过时的引用。logs path 并在行为看起来可疑时检查应用日志。每个屏幕的有用命令:
agent-device --session {SESSION} snapshot -i
agent-device --session {SESSION} screenshot {OUTPUT_DIR}/screenshots/{screen-name}.png
agent-device --session {SESSION} appstate
agent-device --session {SESSION} logs path
一次性完成探索和记录。当发现问题时,停止并完全捕获证据,然后再继续。
使用视频 + 步骤截图:
agent-device --session {SESSION} record start {OUTPUT_DIR}/videos/issue-{NNN}-repro.mp4
agent-device --session {SESSION} screenshot {OUTPUT_DIR}/screenshots/issue-{NNN}-step-1.png
sleep 1
# 执行操作
sleep 1
agent-device --session {SESSION} screenshot {OUTPUT_DIR}/screenshots/issue-{NNN}-step-2.png
3. 捕获最终的损坏状态:
sleep 2
agent-device --session {SESSION} screenshot {OUTPUT_DIR}/screenshots/issue-{NNN}-result.png
4. 停止录制:
agent-device --session {SESSION} record stop
5. 立即将问题附加到报告中,并附上编号步骤和截图引用。
单张截图即可;不需要视频:
agent-device --session {SESSION} screenshot {OUTPUT_DIR}/screenshots/issue-{NNN}.png
在报告中将 复现视频 设置为 N/A。
目标是 5-10 个有充分证据的问题,然后完成:
report.md 中核对摘要严重性计数。agent-device --session {SESSION} close
@eN) 进行快速探索,在需要确定性重放断言时使用选择器。fill 实现清空后输入的语义;使用 type 检查增量输入行为。| 参考 | 何时阅读 |
|---|---|
| references/issue-taxonomy.md | 会话开始时;严重性/类别/检查清单 |
| 模板 | 用途 |
|---|---|
| templates/dogfood-report-template.md | 复制到输出目录作为报告文件 |
每周安装次数
82
仓库
GitHub 星标数
995
首次出现
8 天前
安全审计
安装于
claude-code74
opencode72
github-copilot72
amp72
cline72
codex72
Systematically explore a mobile app, find issues, and produce a report with full reproduction evidence for every finding.
Only the Target app is required. Everything else has sensible defaults.
| Parameter | Default | Example override |
|---|---|---|
| Target app | (required) | Settings, com.example.app, deep link URL |
| Platform | Infer from user context; otherwise ask (ios or android) | --platform ios |
| Session name | Slugified app/platform (for example settings-ios) | --session my-session |
| Output directory | ./dogfood-output/ | Output directory: /tmp/mobile-qa |
| Scope | Full app | Focus on onboarding and profile |
| Authentication | None | Sign in to user@example.com |
If the user gives enough context to start, begin immediately with defaults. Ask follow-up only when a required detail is missing (for example platform or credentials).
Prefer direct agent-device binary when available.
1. Initialize Set up session, output dirs, report file
2. Launch/Auth Open app and sign in if needed
3. Orient Capture initial snapshot and map navigation
4. Explore Systematically test flows and states
5. Document Record reproducible evidence per issue
6. Wrap up Reconcile summary, close session
mkdir -p {OUTPUT_DIR}/screenshots {OUTPUT_DIR}/videos
cp {SKILL_DIR}/templates/dogfood-report-template.md {OUTPUT_DIR}/report.md
Start a named session and launch target app:
agent-device --session {SESSION} open {TARGET_APP} --platform {PLATFORM}
agent-device --session {SESSION} snapshot -i
If login is required:
agent-device --session {SESSION} snapshot -i
agent-device --session {SESSION} fill @e1 "{EMAIL}"
agent-device --session {SESSION} fill @e2 "{PASSWORD}"
agent-device --session {SESSION} press @e3
agent-device --session {SESSION} wait 1000
agent-device --session {SESSION} snapshot -i
For OTP/email codes: ask the user, wait for input, then continue.
Capture initial evidence and navigation anchors:
agent-device --session {SESSION} screenshot {OUTPUT_DIR}/screenshots/initial.png
agent-device --session {SESSION} snapshot -i
Map top-level navigation, tabs, and key workflows before deep testing.
Read references/issue-taxonomy.md for severity/category calibration.
Strategy:
diff snapshot -i after UI transitions to avoid stale refs.logs path and inspect the app log when behavior looks suspicious.Useful commands per screen:
agent-device --session {SESSION} snapshot -i
agent-device --session {SESSION} screenshot {OUTPUT_DIR}/screenshots/{screen-name}.png
agent-device --session {SESSION} appstate
agent-device --session {SESSION} logs path
Explore and document in one pass. When you find an issue, stop and fully capture evidence before continuing.
Use video + step screenshots:
agent-device --session {SESSION} record start {OUTPUT_DIR}/videos/issue-{NNN}-repro.mp4
agent-device --session {SESSION} screenshot {OUTPUT_DIR}/screenshots/issue-{NNN}-step-1.png
sleep 1
# perform action
sleep 1
agent-device --session {SESSION} screenshot {OUTPUT_DIR}/screenshots/issue-{NNN}-step-2.png
3. Capture final broken state:
sleep 2
agent-device --session {SESSION} screenshot {OUTPUT_DIR}/screenshots/issue-{NNN}-result.png
4. Stop recording:
agent-device --session {SESSION} record stop
5. Append issue immediately to report with numbered steps and screenshot references.
Single screenshot is sufficient; no video required:
agent-device --session {SESSION} screenshot {OUTPUT_DIR}/screenshots/issue-{NNN}.png
Set Repro Video to N/A in the report.
Target 5-10 well-evidenced issues, then finish:
report.md.agent-device --session {SESSION} close
@eN) for fast exploration, selectors for deterministic replay assertions when needed.fill for clear-then-type semantics; use type for incremental typing behavior checks.| Reference | When to Read |
|---|---|
| references/issue-taxonomy.md | Start of session; severity/categories/checklist |
| Template | Purpose |
|---|---|
| templates/dogfood-report-template.md | Copy into output directory as the report file |
Weekly Installs
82
Repository
GitHub Stars
995
First Seen
8 days ago
Security Audits
Gen Agent Trust HubPassSocketPassSnykFail
Installed on
claude-code74
opencode72
github-copilot72
amp72
cline72
codex72
Skills CLI 使用指南:AI Agent 技能包管理器安装与管理教程
27,400 周安装
营销活动执行指南:规划、跨渠道执行与效果分析全流程
234 周安装
Google Gemini CLI PR 评论处理助手 - 自动化审查 GitHub Pull Request 反馈
234 周安装
Reddit营销技能:AI生成符合社区规范的Reddit帖子与评论文案工具
234 周安装
sdd:create-ideas - 上下文工程套件中的创意生成AI工具,提供高概率与多样性回复
234 周安装
Vitest 测试框架最佳实践指南:断言、模拟与异步测试完整教程
234 周安装
shinkoku 确定申告能力一览:支持个人事业主、副业、医疗费控除等申报模式
234 周安装