design-harness

Design Harness turns papers, repositories, notes, and rough ideas into a traceable system design.

The agent files sources, connects evidence, records conflicts, and assembles the final output. Human judgment remains in control of every design decision.

The canvas — sources, ideas, and output as three worlds on one board

▶ Open the live canvas — this repo's own design workspace, dogfooded.

Why it exists

Research often spreads across browser tabs, documents, bookmarks, and long chat histories.

The final design may look complete, but the reasoning behind it is easily lost:

Why was this approach chosen?

Design Harness keeps that answer with the project.

ToolWhat it keeps
Reference managerSaved material
Notes appWritten observations
WhiteboardConnections between ideas
Design HarnessThe path from source to idea to decision

How it works

Talk to your agent in plain language:

  1. Drop material — the agent files and grades each source as a card.

  2. Think out loud — your judgments land as idea cards, linked to their evidence.

  3. Say go — the ideas assemble into the output your target.md declares.

  4. Ask for the canvas — the agent builds the board and hands you a clickable link.

Everything is plain Markdown in three folders of your repo — versionable, greppable, no lock-in. The canvas is one self-contained HTML file: commit it, and whoever gets the repo gets the whole board, every card one click from the evidence that earned it.

Get started

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 talk to your agent:

file these papers onto the board
assemble the design

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

More in the works — built in the open, shipped fast.

Explore Tigerless Labs