agentleFS
Sign inStart free
witnesstodarkwitnesstodark/mr-mak-workspace/.agents/skills/workspace-authoring/SKILL.md

workspace-authoring

Skill718 starsChanged 22 days agoMITView on GitHub

Create or maintain Workspace cards, report tabs and Markdown deliverables with consistent styling, working media previews and clear project organization.

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 5 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read AGENTS.md and the workflow.

    line 8
  2. Distinctive:

    Keep authored content in English unless the user explicitly requests another language.

    line 9
  3. Very specific:

    Use workspace/_shared/report.css and report.js in every HTML report, with data-mak-report="document" on the root element.

    line 22
  4. Distinctive:

    Every image must be clickable, preview at full size and offer a download.

    line 31
  5. Distinctive:

    Use local output paths, meaningful alt text and lazy loading.

    line 32
  6. Very specific:

    Update workspace/workspace.json with a stable ID, category, folder, readable title, creation date, last updated date and steps.

    line 36
  7. Distinctive:

    Test at 390 px and 768 px as well as desktop width.

    line 47
  8. Very specific:

    Use relative assets inside the report folder or workspace/_shared/.

    line 50

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Workspace authoring

Read AGENTS.md and the workflow. Keep authored content in English unless the user explicitly requests another language. Put related iterations in an existing card rather than creating several cards for one task. Do not reorganize unrelated projects.

For a gallery or research page that the owner wants to edit directly, use the versioned .card.json format in editable cards. Register its step with viewer: "card". Keep stable entry IDs, relative local images and Markdown notes; validate with desktop/service/card-schema.mjs before saving. Reread the source before each edit and update the registry date. The app provides the view, so do not create a second editable HTML copy of the same data. Preserve existing custom HTML unless conversion was requested. Browser and phone previews can read these cards; the desktop Workspace provides editing controls.

Use workspace/_shared/report.css and report.js in every HTML report, with data-mak-report="document" on the root element. Use the shared tokens and layout instead of inventing another visual theme. Basic cards and media grids can use the sample CSS in workspace/_shared/examples.css. Standard reports follow the reader's Dark, Light or System preference. Check both dark and light when adding custom styles. An intentionally independent design, such as a game demo, can keep its palette with data-mak-theme="custom" on the root element. Do not invert or recolor media to implement a theme.

Every image must be clickable, preview at full size and offer a download. Use local output paths, meaningful alt text and lazy loading. Keep motion studies playable with native video controls. Wide tables must scroll within the report. Never cover the document close button with the Files rail.

Update workspace/workspace.json with a stable ID, category, folder, readable title, creation date, last updated date and steps. sample: true is reserved for starter examples; ordinary work uses normal archiving behavior.

Open every changed tab at wide and narrow sizes. Check media and links, Markdown rendering and image close/download controls. Preserve the selected surface during loading. Report actual verification and any remaining limitations. Prepare desktop changes without restarting active sessions until the user permits it.

Mobile reports use the same HTML or Markdown as desktop. Include the viewport meta tag and shared styles; keep the reading order useful in a single column. Test at 390 px and 768 px as well as desktop width. Verify readable text without zooming, wrapping tab labels, reachable close/download buttons, and working image previews with touch. Confine horizontal scrolling to tables and code. Use relative assets inside the report folder or workspace/_shared/. Reports opened through Mobile Access run in an isolated, read-only preview: do not depend on the desktop control API, parent-window DOM, local storage or absolute localhost URLs. Keep a short result summary before lengthy details.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free