auditor
Skill155 starsChanged 4 months agoMITView on GitHub
Detect and fix layout issues in generated HTML using visual rendering tests. Fixes overlap, spacing, and positioning problems without changing content or design intent.
5 other projects use a version of this.
Adopted by
Most recently taken up by
- NotRedFox/NotRedFoxs-Claude-skillstheir own version, first seen Oct 9, 2026
- opentensor/subtensortheir own version, first seen Oct 8, 2026
- edisonzerolam/openclaw-skillstheir own version, first seen Jul 4, 2026
- daemon-blockint-tech/Agentic-Enteprises-Skilltheir own version, first seen May 20, 2026
- rade-nugroho/Agentic-Enteprises-Skilltheir own version, first seen May 20, 2026
Other versions5 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 5
- opentensor/subtensor★ 402same name
- rade-nugroho/Agentic-Enteprises-Skill★ 9same name
- daemon-blockint-tech/Agentic-Enteprises-Skill★ 8same name
- NotRedFox/NotRedFoxs-Claude-skills★ 1same name
- edisonzerolam/openclaw-skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository28 other files it gives its agents.
- analystSkill
- designerSkill
- detectiveSkill
- editorSkill
- inspectorSkill
- analystSkill
- auditorSkill
- cinematographerSkill
- copywriterSkill
- criticSkill
- designerSkill
- detectiveSkill
- editorSkill
- programmerSkill
- heroSkill
- ideationSkill
- imagineerSkill
- inspectorSkill
- interactionSkill
- programmerSkill
- scoutSkill
- data2story-proSkill
- data2storySkill
- dataviz-craftSkill
- find-dataSkill
- frontend-design-proSkill
- frontend-designSkill
- sparring-partnerSkill
The file
We split this skill into the rules it gives an agent
3 rules. 0 are common, 1 is distinctive, and 2 are very specific to this project.
0
Common
Most of the 4 edited versions we read keep these.
1
Distinctive
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
- Very specific: line 34
After making changes: confirm all data-* attributes are still present, no content text was modified, all asset paths are unchanged, and no …
- Distinctive: line 43
If a chart is blank because the mount element has no width, you may fix the mount container styles — but do not rewrite the Vega data/spec …
- Very specific: line 47
Write PROJECT_DIR/auditor.json — a structured record of what you fixed (an array of {type, count, lines}.
“Common” is measured: kept by at least 6 in 10 of the 4 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Auditor
Your job is layout repair. The programmer has created the HTML with all the right content and visuals, but there may be CSS/layout issues causing overlap, misalignment, or spacing problems. You fix these technical issues without changing the creative intent.
Setup
PROJECT_DIR= first argument- Required file:
PROJECT_DIR/index.html - You will read, analyze, and edit the HTML to fix layout issues
What You Fix
In scope: visual elements overlapping with text; missing wrapper containers around images/videos/charts; incorrect or missing spacing; elements not properly centered or aligned; responsive layout breaking on mobile widths; float/positioning issues causing text wrap; overflow causing horizontal scroll.
Out of scope (do NOT change): content or prose text; visual design choices (colors, fonts, sizes); chart data or specifications; interactive functionality; asset paths or data-* attributes.
Step 1: Detect Issues
Since you cannot run a real browser, use code analysis to detect common layout issues — unwrapped visual elements, missing spacing, centering problems, and responsive gaps. The grep commands and the exact patterns for each check are in references/checks.json.
Step 2: Apply Fixes
For each issue found, use the Edit tool to make minimal, surgical changes (prefer inline styles). The before/after templates for the six common fixes — wrap unwrapped visuals, add chart spacing, center images, add responsive constraints, re-center full-bleed breakouts, and give a collapsed Vega mount a measurable width — are in references/fix_patterns.json.
Step 3: Verify Fixes
After making changes: confirm all data-* attributes are still present, no content text was modified, all asset paths are unchanged, and no closing tags were broken. Re-grep for remaining issues (commands in references/fix_patterns.json → verify_after_fixes). Document the specific issues found and the fixes applied.
Rules
- Use inline styles for fixes (don't modify
<style>blocks unless absolutely necessary) - Preserve all attributes — especially
data-des,data-ana,data-edt,id,class - Don't add new visual elements — only fix layout of existing ones
- Don't change content — no text modifications
- Make minimal edits — only fix what's broken; use the Edit tool, not full-section rewrites
- If a chart is blank because the mount element has no width, you may fix the mount container styles — but do not rewrite the Vega data/spec unless the problem is clearly not layout-related
Step 4: Report Fixes
Write PROJECT_DIR/auditor.json — a structured record of what you fixed (an array of {type, count, lines}; empty array [] if no fixes). Use the canonical type names so they match the error case filenames in errors/cases/. The format and the full type list are in references/report_types.json.
Output
Modified PROJECT_DIR/index.html with layout issues fixed, plus PROJECT_DIR/auditor.json with the structured fix report.
Done when all visual elements are properly wrapped and spaced, no elements overlap with text content, responsive constraints are in place, and all original content and attributes are preserved.
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
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.

