Skip to content
Back to search

fastpilot

MarketplaceWorks with Claude CodeGitHub

Test your web, Android, and iOS apps in plain language, from one dashboard. You say what to check; AI runs it on your own computer, shows you every step live, and tells you what passed and what…

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/autobot-lab--fastpilot/badge.svg)](/listings/autobot-lab--fastpilot)

HTML

<a href="/listings/autobot-lab--fastpilot"><img src="/api/listings/autobot-lab--fastpilot/badge.svg" alt="findsafeskills safety rating" /></a>

Install

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

.claude-plugin/marketplace.json · Manifest · 13 lines
  1. {
  2. "name": "fastpilot",
  3. "owner": { "name": "autobot-lab", "url": "https://github.com/autobot-lab" },
  4. "metadata": { "description": "fastpilot, AI-driven QA for web, Android and iOS, as a Claude Code plugin." },
  5. "plugins": [
  6. {
  7. "name": "fastpilot",
  8. "source": "./",
  9. "description": "An MCP server that drives Chrome, emulators and simulators, plus 24 testing skills.",
  10. "category": "testing"
  11. }
  12. ]
  13. }
README.md · README · 188 lines
  1. # fastpilot
  2. **Test your web, Android, and iOS apps in plain language, from one dashboard.**
  3. You say what to check; AI runs it on your own computer, shows you every step live, and tells you
  4. what passed and what failed. Free, open source, no cloud. Payment and delete-account buttons are
  5. blocked from the start.
  6. <img src="docs/images/demo.gif" width="860" alt="A 60-second tour of fastpilot: one sentence in Chat becomes a test card, Save and schedule runs it every 2 hours, Live shows each step on web and on an Android emulator where the Pay button is blocked, and a failing test is explained in plain words.">
  7. *A 60-second tour, sped up where the AI is thinking. Chat turns one sentence into a test, Live shows each step on web and on an Android emulator (the Pay button is blocked), and a failing test says why.*
  8. <img src="docs/images/07-chat-save-test.png" width="860" alt="The fastpilot dashboard: an agent called QA Checkout gets one sentence in Chat and answers with a Save test and schedule card that lists the test steps, a schedule every 2 hours, and the buttons Save and schedule, Save test only and Edit.">
  9. ## Start in 3 steps
  10. You need a Mac (or Linux), Google Chrome, and [Claude Code](https://claude.com/claude-code) signed in.
  11. **1. Install** (once)
  12. ```bash
  13. npm install -g fastpilot
  14. ```
  15. **2. Set up your project** (once per project)
  16. ```bash
  17. cd ~/your-project
  18. fastpilot init
  19. ```
  20. `init` asks before each change; answer `y`. At the end it asks **"Open the dashboard now?"**:
  21. press Enter.
  22. > Using Claude Code? You can also install fastpilot as a plugin (the MCP server and its 24 skills in
  23. > one step): `/plugin marketplace add autobot-lab/fastpilot`, then `/plugin install fastpilot@fastpilot`.
  24. > Use **either** the plugin **or** `fastpilot init` in a project, not both: both together start two
  25. > fastpilot servers. With the plugin, a secret such as `DEMO_PASS` works like this: add its **name** in
  26. > `fastpilot ui` › Settings › Secret, and set its **value** in the environment you start Claude Code
  27. > from (`export DEMO_PASS=…`). The value is never typed into the dashboard or the chat.
  28. > Just want to look around first? Run `fastpilot demo` instead. It makes a demo folder with demo
  29. > pages and example tests, and opens the dashboard on it.
  30. **3. Open the dashboard** (every time)
  31. ```bash
  32. fastpilot ui
  33. ```
  34. Your browser opens the dashboard. Keep the Terminal open while you use it; `Ctrl-C` closes it.
  35. - The link works **once**, for **10 minutes**. If it expired, press `Ctrl-C` and run `fastpilot ui` again.
  36. - The dashboard only works on this computer.
  37. - Open it yourself; do not ask the AI to open it for you.
  38. Everything else happens in the dashboard.
  39. ## Guide: your first test in the dashboard
  40. ### 1. Check your devices: **Device**
  41. Shows whether Chrome, Android, and iOS are ready, with a ✓ or the reason why not.
  42. - **Install Chrome** and **Set up an emulator** are one-click **Fix** buttons. Each asks you first,
  43. showing the size and the time it takes.
  44. - **iPhone** testing needs a one-time setup with your Xcode (about 5 minutes).
  45. - Only web? Chrome is enough.
  46. <img src="docs/images/02-device.png" width="860" alt="fastpilot's Device page: two Android emulators installed on this computer, each with a Start button, and the iPhone simulators tests can use.">
  47. ### 2. Pick the AI and store passwords: **Settings**
  48. - **AI model:** pick the provider, then the model. Claude Code is the default. Other providers
  49. (OpenAI, OpenRouter, Groq, Ollama) also work, and are marked "not tested yet".
  50. - **Secret:** add the names of passwords your tests need, for example `TEST_PASS`. Tests write them as
  51. `${TEST_PASS}`, and they always show as `***`. Never type a password in Chat.
  52. - **Limits & permissions:** see the next section, "Safe from the start".
  53. ### 3. Make an agent: **Agents › New agent**
  54. An agent is a tester with its own job, devices, instructions, and skills.
  55. - Fill in three things: what it tests, where (Web, Android emulator, iPhone simulator…), and its
  56. instructions. Click a suggested instruction to add it.
  57. - **Skills** are ready-made test know-how, grouped by technique: boundary values, decision tables,
  58. regression, safe checkout, and more. Suggested skills are already ticked, each with a reason.
  59. With Atlassian's MCP server in Claude Code, three Jira skills turn tickets into tests, log a
  60. failed run as a bug and post results; each write to Jira waits for your OK
  61. ([setup](docs/guide.md#jira)).
  62. - Or simply describe the agent in one sentence in Chat and click **Create**.
  63. You can have several agents, for example "Checkout" on web and "Login" on Android. Each has its own
  64. **Chat, Live, Schedules, Instructions, Skills, History** and **Agent settings**.
  65. ### 4. Ask for a test: **Chat**
  66. Write what you want in plain words:
  67. | You want to | Type in Chat |
  68. |---|---|
  69. | Check something now | *Check that login on https://yourapp.com goes to the Dashboard* |
  70. | Save a test | *Create a test for the checkout flow and save it* |
  71. | Test a field properly | *Test the quantity field properly* (it picks the technique and shows what it covered) |
  72. | Run everything | *Run a regression* (newly failing tests come first) |
  73. | Before a release | *Is staging ready to ship?* (one verdict: SHIP, SHIP WITH FIXES, DO NOT SHIP) |
  74. | Understand a failure | *Why did the checkout test fail?* |
  75. When a test is ready, a card shows its steps: **Save & schedule** or **Save test only**. Nothing is
  76. saved before you click. The picture at the top of this page shows such a card.
  77. ### 5. Watch it run: **Live**
  78. Watch every test step by step, on web, Android, and iOS. **Take over** if you need to tap something
  79. yourself, then hand it back to the AI. Several agents can run at the same time; a phone runs one
  80. test at a time and the others wait their turn.
  81. <img src="docs/images/08-live.png" width="860" alt="The Live page: the QA Checkout agent is testing the event tickets page in Chrome; four steps are done and the fifth, waiting for the reservation message, is running, with Enlarge and Take over buttons.">
  82. ### 6. Run it automatically: **Schedules**
  83. - **Every N hours** (with active hours, for example 08:00–20:00), every day, or every week.
  84. - **Run now** for any schedule.
  85. - **Release watch:** after a release, re-check every few minutes for a few hours and only tell you
  86. about new problems.
  87. - Get notified by Slack webhook, Mac notification, or GitHub issue when something fails.
  88. - Or just ask in Chat: *"run the checkout tests every 2 hours"*.
  89. <img src="docs/images/09-schedules.png" width="860" alt="The Schedules page: a test that runs every 2 hours from 08:00 to 20:00 and a release watch that checks every 10 minutes, each with its next run, a passed last run and a Mac notification on failure.">
  90. ### 7. See results: **History** and **Memory**
  91. - **History:** every run and every Chat conversation, with report, screenshots, and video. Delete
  92. one, several, or all, or let them be deleted after 7, 30, or 90 days.
  93. - **Memory:** how each test is doing (Reliable, Unstable, Failing, Set aside), plus **project notes**
  94. every agent reads, for example *"Staging is https://staging.yourapp.com"* or *"Login uses a code"*.
  95. ## Dashboard map
  96. | Menu | What it's for |
  97. |---|---|
  98. | **Agents** | Your testers: Chat, Live, Schedules, Instructions, Skills, History, Agent settings. Also the skill **Library** |
  99. | **Live** | Every running test, from all agents |
  100. | **Schedules** | All schedules and release watches |
  101. | **Memory** | Test health and project notes |
  102. | **History** | Past runs and conversations |
  103. | **Device** | Chrome, Android, iOS readiness and Fix buttons |
  104. | **Settings** | AI model, Limits & permissions, Secret |
  105. ## Safe from the start
  106. - Buttons like **pay, buy, checkout, transfer, delete account** are always blocked.
  107. - Plain **delete/remove** buttons must be confirmed first.
  108. - The AI can make the rules stricter, never looser. To loosen one, **you** approve it: Chat shows a red
  109. card that quotes what you asked, and allowing **Pay** also needs you to type "Pay". Text on a tested
  110. page can never ask for a change.
  111. - Every change is kept in `~/.fastpilot/audit.jsonl`.
  112. Want to test a checkout for real? Ask in Chat: *"Allow the Pay button for this project"*, or open
  113. **Settings › Limits & permissions**.
  114. These limits protect you from what a tested page tells the AI. They are not a lock against programs
  115. on your own computer: anything that runs as you can edit `~/.fastpilot/config.toml`.
  116. ## Prefer the terminal?
  117. Everything also works from Claude Code. Open Claude in your project folder (`claude`) and say it in
  118. plain words, for example *"fastpilot, run a regression"* or *"fastpilot, test the login page"*.
  119. Results, schedules, and history show up in the dashboard too.
  120. In GitHub Actions, use [`autobot-lab/fastpilot-action@v1`](https://github.com/autobot-lab/fastpilot-action):
  121. it runs your tests on every pull request and adds the JUnit and HTML reports.
  122. ## If you get stuck
  123. Open **Device** in the dashboard, or run:
  124. ```bash
  125. fastpilot doctor
  126. ```
  127. | Problem | Try |
  128. |---|---|
  129. | The dashboard link doesn't work | It works once, for 10 minutes. Press `Ctrl-C`, then run `fastpilot ui` again |
  130. | Chrome or Android isn't ready | **Device** › Fix buttons, or `fastpilot setup web` / `fastpilot setup android` |
  131. | iPhone isn't ready | `fastpilot setup ios` (once, with Xcode) |
  132. | fastpilot doesn't show up in Claude | Run `fastpilot init` again |
  133. | A test stops with "Blocked by limits" | That's on purpose. See **Settings › Limits & permissions** |
  134. | A saved test can't find its password in `fastpilot run` | Export it first, e.g. `export FASTPILOT_VARS=TEST_PASS TEST_PASS=...` |
  135. ## Learn more
  136. - **[Full guide](docs/guide.md)**: writing tests by hand, all settings, Android, iOS, Claude Desktop,
  137. devices on another computer, known limits
  138. - [Physical phones](docs/real-devices.md)
  139. - [Release notes](https://github.com/autobot-lab/fastpilot/releases)
  140. License: [Apache-2.0](LICENSE)

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.