Skip to content
Back to search

decision-form

MarketplaceWorks with Claude CodeGitHub

Clickable HTML decision forms for coding agents.

Safety scan

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

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/martinandrt--decision-form/badge.svg)](/listings/martinandrt--decision-form)

HTML

<a href="/listings/martinandrt--decision-form"><img src="/api/listings/martinandrt--decision-form/badge.svg" alt="findsafeskills safety rating" /></a>

Install

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

/plugin marketplace add martinandrt/decision-form
/plugin install decision-form@decision-form

That's it. Next time your agent has three or more decisions for you, or is about to do something hard to undo, it builds a form instead of a wall of text. You can also just say *"make me a decision form for this"*.

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 · 12 lines
  1. {
  2. "name": "decision-form",
  3. "owner": { "name": "Martin Andrt" },
  4. "description": "Clickable HTML decision forms for coding agents.",
  5. "plugins": [
  6. {
  7. "name": "decision-form",
  8. "source": "./",
  9. "description": "Turn a pile of decisions into one tappable page; the agent waits for your answers before acting."
  10. }
  11. ]
  12. }
README.md · README · 176 lines
  1. <p align="center">
  2. <img src=".claude-plugin/icon.png" width="88" alt="">
  3. </p>
  4. <h1 align="center">decision-form</h1>
  5. <p align="center">
  6. <b>Stop answering seven questions buried in one chat message.</b><br>
  7. Your coding agent turns its open decisions into one page of tappable cards, on your laptop or your phone,
  8. and waits for your answers before it touches anything.
  9. </p>
  10. <p align="center">
  11. <img src="https://img.shields.io/badge/Claude_Code-plugin-1d6e77" alt="Claude Code plugin">
  12. <img src="https://img.shields.io/badge/Agent_Skill-SKILL.md-1d6e77" alt="Agent Skill">
  13. <img src="https://img.shields.io/badge/python-3.10%2B-1d6e77" alt="Python 3.10+">
  14. <img src="https://img.shields.io/badge/dependencies-none-1d6e77" alt="No dependencies">
  15. <img src="https://img.shields.io/badge/license-MIT-1d6e77" alt="MIT license">
  16. </p>
  17. <p align="center">
  18. <img src="docs/demo.gif" width="340" alt="Filling a decision form on a phone: tap options, expand an explanation, type a custom answer, send">
  19. </p>
  20. ## Install in two lines
  21. ```text
  22. /plugin marketplace add martinandrt/decision-form
  23. /plugin install decision-form@decision-form
  24. ```
  25. That's it. Next time your agent has three or more decisions for you, or is about to do something hard to undo,
  26. it builds a form instead of a wall of text. You can also just say *"make me a decision form for this"*.
  27. ## What you get
  28. - **One card per decision.** The question, one line on what is at stake (including the downside), the agent's
  29. recommendation marked, an "Explain" fold for the details, and "Other" with a text field on every card.
  30. - **The agent actually waits.** A tiny local server runs in the background; when you tap *Send answers* it saves
  31. `answers.json` and exits, and the agent picks it up on its own. No copy-paste, no polling.
  32. - **Answer from your phone.** `--lan` gives you a link for your phone on the same Wi-Fi. Decide on the couch,
  33. the agent continues at the desk.
  34. - **Works without the server too.** Open the HTML file, tap *Copy answers*, paste `1a 2c 3x: ask QA first`
  35. into the chat.
  36. - **A decision record for free.** The questions (`.md`), the page (`.html`) and your answers (`.answers.json`)
  37. stay side by side in `decisions/`.
  38. - **Nothing leaves your machine.** One self-contained HTML file, no CDN, no fonts to fetch, no tracking. Python
  39. standard library only. Light and dark mode.
  40. ## How it works
  41. ```mermaid
  42. sequenceDiagram
  43. participant A as Agent
  44. participant F as Form page
  45. participant Y as You
  46. A->>A: write decisions/2026-11-17-cleanup.md, build the page
  47. A->>F: start serve_form.py in the background
  48. A-->>Y: here is the link (laptop or phone)
  49. Y->>F: tap through the cards, Send answers
  50. F->>A: answers.json saved, server exits
  51. A-->>Y: recap of what it will do, then does it, safest wave first
  52. ```
  53. Unanswered questions are asked about, never filled in with the agent's own guess.
  54. ## A card, as the agent wrote it
  55. Output of a fresh agent (Claude Haiku) that had only this skill and the task *"decide language, license, CI
  56. and registry for a new CLI"*. Only the line breaks are ours:
  57. ```markdown
  58. ### Which license: MIT or Apache-2.0?
  59. Both let anyone use, change and sell the code. Changing the license is hard
  60. once outside contributors have sent code, so decide before that happens.
  61. Downside of MIT: it says nothing about patents.
  62. > MIT is one short paragraph and the most common license for small tools.
  63. > Apache-2.0 is longer. It adds an explicit patent license from every
  64. > contributor and ends that license for anyone who sues over the code.
  65. > It cannot be combined with GPLv2 code, which rarely matters for a new CLI.
  66. - MIT: short, and the most common for small tools *
  67. - Apache-2.0: explicit patent grant, better if companies will contribute
  68. ```
  69. The skill teaches the agent how to write these: plain words instead of jargon, the downside next to the benefit,
  70. exactly one recommendation, and reversible decisions before irreversible ones.
  71. ## Why not just chat, or the built-in question tool?
  72. | | Wall of text in chat | Built-in question tool (AskUserQuestion) | decision-form |
  73. | ------------------------------ | -------------------- | ---------------------------------------- | ------------- |
  74. | Questions at once | any, hard to track | 1 to 4 | any, grouped in waves |
  75. | Agent's recommendation | buried in prose | by convention | on every card |
  76. | What is at stake | mixed in | short description | one line + "Explain" |
  77. | Answer from your phone | no | no | yes |
  78. | Record of what was decided | scrollback | scrollback | `.md` + `.answers.json` |
  79. For one or two quick questions, the built-in tool is the right choice. The skill says so too.
  80. ## Try it without an agent
  81. ```bash
  82. git clone https://github.com/martinandrt/decision-form && cd decision-form
  83. python3 skills/decision-form/scripts/serve_form.py examples/repo-cleanup.html --open
  84. ```
  85. Tap through, send, and the answers print in your terminal:
  86. ```text
  87. Answers: 1a 2a 3b 4a 5a 6x: Ask the QA team first 7a
  88. 1. Delete 23 merged branches? -> a) Delete all 23
  89. 2. Move CI from Travis to GitHub Actions? -> a) Port all four jobs
  90. 3. Drop Node 18 support? -> b) Keep it one more release with a warning
  91. ...
  92. Saved: examples/repo-cleanup.answers.json
  93. ```
  94. ## Other ways to install
  95. **Copy the skill** (personal skills, all projects):
  96. ```bash
  97. git clone https://github.com/martinandrt/decision-form
  98. cp -R decision-form/skills/decision-form ~/.claude/skills/
  99. ```
  100. From the shell instead of a session: `claude plugin marketplace add martinandrt/decision-form` and
  101. `claude plugin install decision-form@decision-form`.
  102. **Other agents** that read Agent Skills: copy `skills/decision-form` into the folder your agent loads skills
  103. from (for example `~/.agents/skills/` or `.cursor/skills/`). Only Claude Code is tested so far.
  104. ## Write a form by hand
  105. ```markdown
  106. # Repo cleanup before v2.0
  107. Seven decisions, safest first. Tap a choice on each card, then send.
  108. ## Wave 1 · safe, reversible
  109. ### Delete 23 merged branches?
  110. All of them are merged into main and none has an open pull request.
  111. > Longer explanation, hidden behind "Explain".
  112. - Delete all 23 *
  113. - Keep release/* branches, delete the rest
  114. - Keep some of them (write which) [write]
  115. ```
  116. ```bash
  117. python3 skills/decision-form/scripts/build_form.py my-decisions.md # -> my-decisions.html
  118. ```
  119. `#` title · line under it = intro · `##` group · `###` question · plain line = what is at stake · `>` = explanation ·
  120. `- option` · trailing `*` = recommended · `[write]` = text field. JSON input works too (`--syntax` shows both).
  121. Czech interface: `--lang cs`.
  122. <p align="center"><img src="docs/light-dark.png" width="520" alt="The same form in light and dark mode"></p>
  123. ## Privacy and safety
  124. - The page makes no network requests; your answers stay on your machine.
  125. - The server listens on `127.0.0.1` by default and every request needs the random token from the printed link.
  126. - `--lan` opens it to your local network over plain HTTP. Use it only on a network you trust.
  127. - The server accepts one submission and exits. With no answer in 60 minutes (`--timeout`) it exits with code 2.
  128. - Tapping *Send* with questions still open asks for a second tap, so a stray tap can't send half a form.
  129. ## Background
  130. I built this for my own daily work with Claude Code: about 130 forms since August 2026, for cleanups,
  131. refactors, architecture calls and anything I didn't want decided by accident. The card-writing rules in
  132. [`SKILL.md`](skills/decision-form/SKILL.md) come from that use.
  133. Tests: `python3 -m unittest discover tests` (the server tests need to bind a local port).
  134. ## License
  135. MIT, see [LICENSE](LICENSE).

Ratings and comments

Sign in to rate this skill.

Sign in

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