agentleFS
Sign inStart free
imMamdouhaboammarimMamdouhaboammar/linkedin-animated-infographics/skills/qa-post/SKILL.md

qa-post

Skill14 starsChanged 3 months agoMITView on GitHub

Run deterministic render checks, local creative gates, adversarial critique, and independent evidence-backed acceptance against a built infographic before shipment.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 2 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    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. Very specific:

    Read helper/GUIDE.md before checking the artifact.

    line 16
  2. Distinctive:

    Use for a finished or nearly finished artboard/GIF that needs shipping acceptance, regression review, visual/caption red-teaming, or …

    line 20
  3. Distinctive:

    Run deterministic render checks:

    line 36
  4. Very specific:

    If a caption exists, apply the caption skill and hooked-design-copy.

    line 45

≈ “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

/linkedin-animated-infographics:qa-post

Arguments: $ARGUMENTS

Purpose

Act as the focused parent workflow for QA. Report evidence and route targeted fixes; do not silently edit the artifact or let the producing worker grade its own result.

Read helper/GUIDE.md before checking the artifact.

Use when

Use for a finished or nearly finished artboard/GIF that needs shipping acceptance, regression review, visual/caption red-teaming, or verification after a targeted fix.

Inputs

  • artboard HTML and rendered static/GIF evidence
  • optional caption and first comment
  • story brief, selected creative concept, evidence table, and layout spec when available
  • optional mascot identity/motion contract
  • explicit acceptance criteria when supplied

Outputs

Return build/render-report.json, build/critic-report.json, build/verification-report.json, and one final verdict: SHIP or HOLD: <the one thing to fix first>.

Procedure

  1. Run deterministic render checks:
bash ${CLAUDE_PLUGIN_ROOT}/scripts/lint_artboard.sh <path>
python3 ${CLAUDE_PLUGIN_ROOT}/scripts/artboard_audit.py <path> --json /tmp/artboard-audit.json
python3 ${CLAUDE_PLUGIN_ROOT}/scripts/check_render.py <path> --out /tmp/qa-still.png --mobile --json /tmp/still-audit.json
  1. Walk the render skill and references/qa-gates.md. Inspect frame 0 and feed-width output. For GIFs, use render.sh so lint, measured artboard/still/GIF fragments, and the merged build/render-report.json are all produced.
  2. If a caption exists, apply the caption skill and hooked-design-copy. Check truncation, one archetype, factual support, CTA discipline, and anti-slop findings.
  3. Delegate adversarial review to post-critic. It must evaluate creative-payoff, restrained-palette, center-first-composition, UI fidelity, mascot identity when present, motion meaning, and evidence safety.
  4. Any must-fix item produces a HOLD until corrected or proven not applicable.
  5. Delegate independent acceptance to story-verifier. The verifier reads the actual artifacts and evidence, not a worker summary.
  6. Respect the maximum-two targeted fix attempts before escalation.

HOLD conditions

Return HOLD for a failed render gate, unsupported factual claim, generic/unsupported hero hook, missing promised creative payoff, excessive or unjustified palette treatment, invalid alignment exception, UI/mascot fidelity failure, or independent verifier failure.

Related components

  • routing authority: helper/GUIDE.md
  • local quality gates: helper/quality-gates.json
  • render skill: skills/render/SKILL.md
  • caption skill: skills/caption/SKILL.md
  • adversarial worker: agents/post-critic.md
  • independent verifier: agents/story-verifier.md
  • verification reference: skills/info-stories/references/verification-loop.md

Research gates

QA enforces applicable prose-specificity, voice-preservation, structural-originality, contrast-discipline, and evidence-traceability gates. bounded-verification is always active for final acceptance and prevents unbounded repair loops.

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