agentleFS
sreejithramansreejithraman/SreeStack/skills/showroom/SKILL.md

showroom

Skill1 starsChanged 16 days agoMITView on GitHub

Use when work produces something the user can inspect or try, or during UI iteration to show progress. Choose and deliver demonstrations and access suited to the work, the user's next step, and available tools.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read the matching delivery guidance before selecting access.

    line 25
  2. Distinctive:

    Use the project's run path and the host's browser, device, or terminal routing.

    line 54
  3. Distinctive:

    When a chosen route is unavailable, use another suitable format if it still helps, and label the gap.

    line 64
  4. Distinctive:

    Keep services needed for the user's inspection running, and record ownership and exact stop actions as described in delivery.

    line 77

≈ “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

Showroom

Help the user experience and judge the result. Choose how to show it and how to make it available, using the task, environment, and user's preferences. This applies to interfaces, commands, services, generated files, and other inspectable results.

verify owns acceptance checks and their judgments. Reuse its current evidence and selected environment when they fit; use it when broader workflow checking is needed. Showroom owns the demonstration and access at handoff and during iteration. A demonstration supports only the behavior actually observed.

Choose the showing

Decide what would help the user assess the work and what would let them try it. Inspect existing run artifacts, available tools, running services, and suitable devices before choosing. Follow explicit format and target requests; otherwise make the choice without asking the user to select a medium.

Read the matching delivery guidance before selecting access. Website work always follows its preview-link requirement.

Choose formats for what they communicate. These are examples, not a ranking:

  • Screenshots make appearance and individual states easy to inspect.
  • A collage or contact sheet makes states, variants, or before/after views easy to compare; retain access to the full-size images.
  • Video shows timing, gestures, transitions, and connected interactions.
  • A checked URL lets the user explore a website themselves.
  • A development install puts an app on an available device, including the user's iPhone, when hands-on use is valuable.
  • A runnable command, request and response, or sample session shows nonvisual behavior. An accessible generated file lets the user inspect the artifact.

Combine formats when they answer different questions. During UI iteration, use screenshots or video readily to make feedback concrete, even when a link or installed build also provides access. Choose enough context to judge the change; optimize for understanding rather than artifact count.

A document or configuration edit may be best shown through a direct file link and a focused explanation. Match the showing to the work instead of creating media for its own sake. Reuse suitable evidence while its revision, setup, and state still match. Read captures when producing visual media.

Exercise and inspect

Reach the changed behavior through the real interface or public entry point. Use the project's run path and the host's browser, device, or terminal routing. For web and iOS tools, follow Verify's web and iOS routing; delivery target choices follow delivery.

Confirm the current changes are present. Inspect each artifact as delivered: open the URL or file, view the saved media, or run the example. Use safe sample data and remove secrets or private content from evidence. A loading screen, stale build, home screen, or fabricated output cannot stand in for the result.

When a chosen route is unavailable, use another suitable format if it still helps, and label the gap. Preserve explicit target requirements and distinguish secondary evidence from checks on that target. Report the exact blocker when the result cannot be shown or made accessible.

Handoff

Present the selected media, links, artifacts, or runnable examples in the reply. Explain what the user is seeing and how to reach or try the changed behavior. For images and video, use absolute paths so the host can render them; link generated files directly. Include the full-size sources for composite media.

State relevant access requirements, availability limits, and untested behavior. Keep services needed for the user's inspection running, and record ownership and exact stop actions as described in delivery. Finish when the user can inspect the result and use the offered access path, or the handoff clearly identifies what remains unavailable and why.

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.