Skip to content
Back to search

design-harness

Feed your agent papers and half-formed ideas — it links them into a system design you can defend. Markdown keeps the record; a visual canvas makes it readable. An Agent Skill for Claude Code & any SKILL.md-compatible agent.

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/tigerless-labs--design-harness/badge.svg)](/listings/tigerless-labs--design-harness)

HTML

<a href="/listings/tigerless-labs--design-harness"><img src="/api/listings/tigerless-labs--design-harness/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

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

Claude Code:

/plugin marketplace add tigerless-labs/design-harness
/plugin install design-harness@design-harness

Codex (CLI and the ChatGPT desktop app share one plugin system):

codex plugin marketplace add tigerless-labs/design-harness

then install from `/plugins` in the CLI or the desktop app's plugin directory.

For other SKILL.md-compatible agents, copy the skill folder into your agent's skill directory:

git clone https://github.com/tigerless-labs/design-harness
cp -r design-harness/plugins/design-harness/skills/design-harness \
  ~/.claude/skills/

Updating

Claude Code — refresh the marketplace, then update the plugin:

/plugin marketplace update design-harness
/plugin update design-harness@design-harness

(Or enable auto-update for this marketplace under `/plugin` → Marketplaces.)

Codex — refresh the marketplace snapshot, then update from `/plugins`:

codex plugin marketplace upgrade

Manual copy — `git pull` the clone and re-run the `cp -r` above.

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 · 14 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. {
  2. "name": "design-harness",
  3. "owner": {
  4. "name": "tigerless.ai",
  5. "email": "saas@tigerless.com"
  6. },
  7. "plugins": [
  8. {
  9. "name": "design-harness",
  10. "source": "./plugins/design-harness",
  11. "description": "Step one of vibe coding — turn scattered reading and half-formed ideas into a design you can defend. Three markdown layers (sources → ideas → output) synced on the human's command on a visual canvas; the human adjudicates, the agent runs errands."
  12. }
  13. ]
  14. }
README.md · README · 206 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. <h1 align="center">design-harness</h1>
  2. <p align="center">
  3. <img src="https://img.shields.io/badge/release-v0.10.0-brightgreen.svg" alt="release" /> <img src="https://img.shields.io/badge/agent-Claude%20Code%20%7C%20Codex-blue.svg" alt="agent" /> <img src="https://img.shields.io/badge/format-SKILL.md-lightgrey.svg" alt="format" /> <img src="https://img.shields.io/badge/license-MIT-yellow.svg" alt="license MIT" />
  4. </p>
  5. **Turn scattered reading and half-formed ideas into a design you can defend** —
  6. a skill, a system, a video script, even the structure of an article.
  7. The human adjudicates, the agent runs the errands.
  8. An [Agent Skill](https://agentskills.io) (Claude Code plugin, works with any
  9. SKILL.md-compatible agent): feed it sources and rough ideas as you go — it files
  10. and links them, records how your thinking evolves, and assembles the evidence into
  11. a system design.
  12. Markdown keeps the record; the canvas makes it readable — every source and idea
  13. you add lands on the board, one glance away from any decision's why.
  14. [![The canvas — sources, ideas, and output as three worlds on one board](docs/assets/canvas-styles/hero.png)](https://tigerless-labs.github.io/design-harness/)
  15. <p align="center"><strong><a href="https://tigerless-labs.github.io/design-harness/">▶ Click the board to open the live canvas</a></strong> — this repo's own design workspace, dogfooded — Pages serves the committed <code>docs/canvas.html</code> as-is. It's also our answer to "why is the design like this?" — send the link.</p>
  16. ## What This Does
  17. - **Sources file themselves** — drop papers, repos, blog posts; the agent files one
  18. card per source, grades it, and anchors every claim to where it came from.
  19. - **Ideas are yours alone** — only a human creates an idea card; the agent links it
  20. to evidence, clusters neighbors, and surfaces conflicts instead of picking a side.
  21. - **Output assembles on your word** — one command turns surviving ideas into the
  22. deliverable, and every later sync runs on your word too; edit output directly and
  23. it flows back into ideas.
  24. - **Everything is traceable** — every design element links back through ideas to the
  25. sources that earned it, with an append-only log on every card.
  26. - **Plain Markdown, no lock-in** — three folders in your repo; versionable,
  27. greppable, renders on GitHub. The agent is just the runtime.
  28. - **One visual canvas** — a self-contained HTML projection of the whole board: no
  29. server, no dependencies, rebuilt from the markdown on demand. Commit it, and whoever
  30. gets the repo gets the board.
  31. ## Installation
  32. Claude Code:
  33. ```
  34. /plugin marketplace add tigerless-labs/design-harness
  35. /plugin install design-harness@design-harness
  36. ```
  37. Codex (CLI and the ChatGPT desktop app share one plugin system):
  38. ```
  39. codex plugin marketplace add tigerless-labs/design-harness
  40. ```
  41. then install from `/plugins` in the CLI or the desktop app's plugin directory.
  42. For other SKILL.md-compatible agents, copy the skill folder into your agent's skill
  43. directory:
  44. ```bash
  45. git clone https://github.com/tigerless-labs/design-harness
  46. cp -r design-harness/plugins/design-harness/skills/design-harness \
  47. ~/.claude/skills/
  48. ```
  49. ### Updating
  50. Claude Code — refresh the marketplace, then update the plugin:
  51. ```
  52. /plugin marketplace update design-harness
  53. /plugin update design-harness@design-harness
  54. ```
  55. (Or enable auto-update for this marketplace under `/plugin` → Marketplaces.)
  56. Codex — refresh the marketplace snapshot, then update from `/plugins`:
  57. ```
  58. codex plugin marketplace upgrade
  59. ```
  60. Manual copy — `git pull` the clone and re-run the `cp -r` above.
  61. ## Usage
  62. Talk to your agent in plain language:
  63. ```
  64. > file these papers onto the board
  65. ```
  66. ```
  67. > assemble the design
  68. ```
  69. 1. Drop material — the agent files and grades each source as a card.
  70. 2. Think out loud — your judgments land as idea cards, linked to their evidence.
  71. 3. Say go — the ideas assemble into the output your `target.md` declares.
  72. 4. Ask for the canvas — the agent builds the board and hands you a clickable link.
  73. Render your own workspace in ten seconds — the
  74. [demo board](https://tigerless-labs.github.io/design-harness/) is the same
  75. projection run on a real workspace:
  76. ```bash
  77. python3 plugins/design-harness/skills/design-harness/scripts/build_canvas.py \
  78. path/to/your-workspace -o /tmp/canvas
  79. open /tmp/canvas/canvas.html
  80. ```
  81. ## Share the Why
  82. The board is not just where you think — it's how you show someone else why the design
  83. is what it is. The canvas is one self-contained HTML file, so commit it beside your
  84. workspace:
  85. ```bash
  86. python3 plugins/design-harness/skills/design-harness/scripts/build_canvas.py \
  87. path/to/your-workspace -o docs
  88. git add docs/canvas.html
  89. ```
  90. Whoever gets the repo opens one file and gets the whole board — every card one click
  91. from the evidence that earned it. Want a URL instead? Serve the same file with hosted
  92. pages, the way [this repo does](https://tigerless-labs.github.io/design-harness/).
  93. ## Canvas Styles
  94. Five skins ship with the canvas; switch live from the toolbar, Pin & Paper is the
  95. default. Screenshots below are this repo's own workspace, unretouched — the board
  96. view and a close-up for each style.
  97. ### Pin & Paper (default)
  98. <p>
  99. <img src="docs/assets/canvas-styles/pin-and-paper-1.png" width="49%" alt="Pin & Paper — board">
  100. <img src="docs/assets/canvas-styles/pin-and-paper-2.png" width="49%" alt="Pin & Paper — close-up">
  101. </p>
  102. > A handmade pinboard: yellow paper ground, ink-blue hand script, pinned notes,
  103. > taped slips, and dashed hand-drawn threads between them.
  104. ### Notebook Tabs
  105. <p>
  106. <img src="docs/assets/canvas-styles/notebook-tabs-1.png" width="49%" alt="Notebook Tabs — board">
  107. <img src="docs/assets/canvas-styles/notebook-tabs-2.png" width="49%" alt="Notebook Tabs — close-up">
  108. </p>
  109. > Warm ivory stationery: serif italic headings, index cards with colored tab
  110. > tongues, sticky-note ideas, stitched dashed connections.
  111. ### Swiss Modern
  112. <p>
  113. <img src="docs/assets/canvas-styles/swiss-modern-1.png" width="49%" alt="Swiss Modern — board">
  114. <img src="docs/assets/canvas-styles/swiss-modern-2.png" width="49%" alt="Swiss Modern — close-up">
  115. </p>
  116. > White, ink, and one red accent: hairline grid, giant grotesk world titles with
  117. > red ordinals, ruler-straight edges.
  118. ### BlockFrame
  119. <p>
  120. <img src="docs/assets/canvas-styles/block-frame-1.png" width="49%" alt="BlockFrame — board">
  121. <img src="docs/assets/canvas-styles/block-frame-2.png" width="49%" alt="BlockFrame — close-up">
  122. </p>
  123. > Neo-brutalist pastel decks: thick borders, hard offset shadows, cyan / pink /
  124. > yellow card blocks that tilt just enough to feel alive.
  125. ### 8-Bit Orbit
  126. <p>
  127. <img src="docs/assets/canvas-styles/8-bit-orbit-1.png" width="49%" alt="8-Bit Orbit — board">
  128. <img src="docs/assets/canvas-styles/8-bit-orbit-2.png" width="49%" alt="8-Bit Orbit — close-up">
  129. </p>
  130. > A CRT arcade in deep navy — scanlines, neon cyan / yellow / magenta frames, and
  131. > orthogonal circuit-trace edges. Dark in both color schemes; a CRT has no daylight
  132. > mode.
  133. ## Philosophy
  134. 1. The human adjudicates, the agent runs the errands — agents execute well and judge
  135. poorly, and a wrong design costs the whole branch of code built on it.
  136. 2. Markdown is the only source of truth — every other surface, the canvas included,
  137. is a projection: rebuilt wholesale from the markdown, committed and shared or thrown
  138. away, never the place facts live.
  139. 3. Rejected ideas are archived with their reasons, never deleted — six months later,
  140. "why didn't we do X?" has an answer.
  141. 4. Every element of the output must answer "why?" with a link.
  142. This repo is dogfooded on itself: its own design was produced by running the method
  143. on ~80 sources about decision-making methods.
  144. ## License
  145. MIT — vendored renderers ([marked](https://github.com/markedjs/marked),
  146. [DOMPurify](https://github.com/cure53/DOMPurify)) keep their original headers.
  147. ---
  148. © Tigerless · [tigerless.ai](https://tigerless.ai) · [tigerless.com](https://www.tigerless.com)

Ratings and comments

Sign in to rate this skill.

Sign in

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