agentleFS
thiagoxikotathiagoxikota/figma-maxxing/skills/figma-slop-check/SKILL.md

figma-slop-check

Skill12 starsChanged 42 days agoMITView on GitHub

Review the craft and technical precision of a requested Figma scope with project canon, measures and visual evidence; classify swap/snap/ask while respecting existing correction authorization.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    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:

    Check the property/relationship in question: bindings, style/scale, component integrity, sizing, glossary and real pixels.

    line 34

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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

Figma craft and precision

Audit the requested screen/scope with evidence. Use only the applicable design, copy and handoff criteria; a status reply or documentary typo does not trigger a full canvas audit. Review is read-only; corrections of your own work or an approved fix request use the current authorization, without a new per-item yes.

Applicable sources

  • Project map/glossary/DS and current state govern expected values; project sample-data and locale rules override generic examples.
  • slop-lens.md: overlap/density, process residue, visual excess, repetition, copy and communication criteria.
  • references/checks.md: 11 precision checks and bounded read-only extraction.
  • references/punch-list.md: severity, swap/snap/ask and scoped decisions.
  • figma-canon routes API, state, naming, handoff or anomaly detail when this operation needs it. A numerical rubric can serve a requested scored audit; concrete blockers and evidence determine readiness, not an automatic 8/10 stamp.

Check the property/relationship in question: bindings, style/scale, component integrity, sizing, glossary and real pixels. Reuse valid collection, refresh when live changes/partial failure make it uncertain. Node ids can persist; validate file/target rather than asserting every id is session-specific.

For visual claims, inspect the affected frame/state and critical elements at a useful scale. A whole-section overview proves different things from a crop; no fixed screenshot count proves completeness. Record truncation, variable-mode limits, stale export risk and what was actually inspected.

Findings and action

Each finding carries source, node/path, found/expected, severity and action: swap (identical equivalent), snap (signed measurable distance) or ask (a new design decision). Approved snaps retain their distance in the run’s fixes; intentional/debt decisions use .figma-slop-check/decisions.json with scope and reason. Do not silently erase debt or invent canon.

A requested audit reports findings without applying them. An authorized fix can batch its approved scope, preserving lock/preflight and read-back. "Fix all" does not decide an unresolved product question. New evidence/regression justifies rechecking; no mandatory double pass, hidden checklist log or single-agent-versus-fanout ritual. Stop repeated failures for diagnosis.

Handoff adds its pertinent completeness criteria and can reuse this evidence. Report blockers, warnings and coverage honestly; do not send PR/Figma replies or publish a deliverable merely because a reviewer marked it ready. Runtime memory writes follow the runtime contract, not this review skill.

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.