agentleFS
bohewubohewu/agents_pipeline/skills/artgen-scaffold/SKILL.md

artgen-scaffold

Skill1 starsChanged 4 months agoMITView on GitHub

Write one paste-ready 2D asset prompt or a complete versioned brief and handoff. Use only when the requested deliverable is documentation, not an image.

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 4 are distinctive, and 2 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.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Write a directly usable prompt or a reviewable brief and prompt package for one bounded 2D asset or asset family.

    line 9
  2. Very specific:

    If the request says sprite without mentioning animation, frames, a loop, a cycle, or a sequence, treat it as one static sprite.

    line 11
  3. Distinctive:

    Use prompt-only output only when the user explicitly asks for only a prompt, one paste-ready prompt, or no brief or handoff package, and …

    line 21
  4. Very specific:

    Put every necessary inference visibly inside the block as Assumption: ..., especially a missing size or style.

    line 27
  5. Distinctive:

    Do not create files or claim that an asset was generated.

    line 29
  6. Distinctive:

    If prompt-only conflicts with an explicit full machine-readable or structured handoff, clarify only that delivery ambiguity.

    line 33

≈ “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

Artgen Scaffold

Write a directly usable prompt or a reviewable brief and prompt package for one bounded 2D asset or asset family. Pixel art is the canonical example; adjacent 2D styles are supported when style and dimensions are explicit.

If the request says sprite without mentioning animation, frames, a loop, a cycle, or a sequence, treat it as one static sprite.

Boundary

This skill produces documentation only. It does not generate or edit images, call a provider, run post-processing, pack atlases, or promise emitted assets. Do not intercept an ordinary image-generation or editing request merely because it mentions sprites, tiles, or icons.

If the user explicitly invokes this scaffold while requesting an image, explain that the scaffold cannot deliver the image and do not force a full handoff first. The host may handle generation separately only when its capabilities, authorization, and higher-priority rules permit it; this skill does not create a bridge, start another agent, or grant that permission. If an image edit lacks an accessible source image, request the source rather than claiming to have inspected or preserved it.

Choose the Delivery Depth

Use prompt-only output only when the user explicitly asks for only a prompt, one paste-ready prompt, or no brief or handoff package, and does not also require an incompatible structured package.

Prompt only

  • Return exactly one fenced text block containing one self-contained, directly usable prompt. Add no Request Record, External Handoff Package, repeated prompt, or surrounding delivery prose.
  • Include asset type, style, dimensions, subject, viewpoint, palette, background, and necessary consistency constraints.
  • Put every necessary inference visibly inside the block as Assumption: ..., especially a missing size or style.
  • Preserve identifiers, naming, or version constraints supplied by the user, but do not require IDs, version packaging, a directory plan, atlas planning, or manual checks.
  • Do not create files or claim that an asset was generated.

Finish when the single block is self-contained, the requested dimensions and style are explicit, necessary assumptions are visible, and no generation claim or extra package was added.

If prompt-only conflicts with an explicit full machine-readable or structured handoff, clarify only that delivery ambiguity. Do not silently replace an existing export or consumer's full-output contract.

Full handoff

A complete asset brief, versioned asset family, or request for the complete package uses the full handoff. Before preparing it, read references/full-handoff.md and follow its seven-section contract, exact field labels, identifiers, versioning, naming, assumptions, and final checks. Do not produce the full handoff from this summary alone.

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.