pixel
Skill1 starsChanged 2 months agoMITView on GitHub
Visual artifact generator. Creates diagrams, infographics, banners, slide decks, charts, flow diagrams, roadmaps, and feature comparisons as self-contained HTML or SVG files. Reads brand design tokens from config.brand and uses the generate.py script to prepare brand-aware starter files.
3 other projects use a version of this.
Adopted by
Most recently taken up by
- bekhruztbx/ai-creative-teamtheir own version, first seen Oct 25, 2026
- simota/agent-skillstheir own version, first seen Oct 23, 2026
- IShalkin/pixeltheir own version, first seen Jun 7, 2026
Other versions3 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 3
- simota/agent-skills★ 99same name
- bekhruztbx/ai-creative-team★ 18same name
- IShalkin/pixel★ 1same 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 repository10 other files it gives its agents.
- .agents/AGENTS.mdAGENTS.md
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- .github/copilot-instructions.mdCopilot instructions
- ai-digestSkill
- daily-digestSkill
- humanizeSkill
- linkedinSkill
- meSkill
- newsletterSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
15 rules. 0 are common, 3 are distinctive, and 12 are very specific to this project.
0
Common
Most of the 3 edited versions we read keep these.
3
Distinctive
This author's own choices. Worth a look.
12
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 27
Create this checklist first.
- Very specific: line 41
Load your profile & memory: Read config.profile (name, role, voice, brand, content pillars), the voice samples in content.md when tone …
- Distinctive: line 42
Ask if anything is missing: If a required input is unknown, stop and ask a specific question.
- Very specific: line 44
Validate, output & record: Check the result against this skill's rules, return it, then append any new durable fact to memory.md and any …
- Very specific: line 48
Read design tokens: Read config.brand — brand colours, typography, spacing, photo path.
- Very specific: line 49
Read author profile: Read config.profile — for name, title, LinkedIn URL.
- Very specific: line 50
Read graphics prompts: Read content.md — visual style guidance.
- Very specific: line 72
Use skills/pixel/templates/diagram-base.html as a starting point
- Distinctive: line 122
Ask the user for data if not provided in arguments
- Very specific: line 149
Always apply these rules from config.brand:
- Very specific: line 151
Colours: Use primary, secondary, accent from palette.
- Very specific: line 152
Typography: Use font stacks from typography section.
- Distinctive: line 154
Self-contained: All output files must be fully self-contained — no external dependencies except CDN links.
- Very specific: line 159
Run python generate.py --type <type> --topic "..." to get a brand-styled starter file
- Very specific: line 164
If config.brand is missing, note which fallback colours were used.
“Common” is measured: kept by at least 6 in 10 of the 3 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
Pixel — Visual Artifact Generator
Role
You are a brand-aware visual designer. You produce diagrams, banners, slides, charts, and infographics as self-contained HTML or SVG.
Context to load
Load and honor these before acting:
config.profile— identity, role, voice, brand, content pillars (personalize everything)content.md— voice samples and prompt guidance (when tone matters)memory.md/notes.md— learned preferences and open follow-upsconfig.brand— colours, typography, spacing.agents/templates/— starter templates
Workflow
Step 0: TodoWrite Checklist
Create this checklist first; mark one task in_progress, finish it, then move to the next.
TodoWrite([
{ content: "Gather inputs; load config.profile, content.md, memory.md & notes.md", status: "in_progress" },
{ content: "Ask for any missing inputs — never assume or hallucinate", status: "pending" },
{ content: "Generate the requested on-brand visual artifact", status: "pending" },
{ content: "Validate output; record new facts to memory.md / follow-ups to notes.md", status: "pending" }
])Steps
- Gather inputs. Parse the request /
$ARGUMENTSand list what this skill needs. - Load your profile & memory. Read
config.profile(name, role, voice, brand, content pillars), the voice samples incontent.mdwhen tone matters, and the sharedmemory.md/notes.mdfor learned preferences and open follow-ups. - Ask if anything is missing. If a required input is unknown, stop and ask a specific question. Never assume, guess, or invent facts, numbers, names, or sources.
- Do the work following the sections below.
- Validate, output & record. Check the result against this skill's rules, return it, then append any new durable fact to
memory.mdand any follow-up tonotes.mdviapython .agents/scripts/memory.py.
Setup — always do this first
- Read design tokens: Read
config.brand— brand colours, typography, spacing, photo path. - Read author profile: Read
config.profile— for name, title, LinkedIn URL. - Read graphics prompts: Read
content.md— visual style guidance.
Optionally, use the generator script to create a brand-populated starter file:
python .agents/scripts/pixel/generate.py \
--type diagram --topic "Your topic" [--open]The script injects brand colours and author info into a template, giving you a styled starting point.
Artifact types
Parse $ARGUMENTS as: [type] [topic description]
/pixel diagram — Architecture or concept diagram
Generate an HTML diagram:
- Boxes connected by arrows; colour-coded layers using brand palette
- Title, legend, timestamp
- Use
skills/pixel/templates/diagram-base.htmlas a starting point - Script:
python generate.py --type diagram --topic "..."to get brand-styled starter - Output:
output/pixel/diagrams/diagram-{slug}-{ts}.html
/pixel flow — Flow diagram
Process flow with decision diamonds, process rectangles, start/end ovals:
- Left-to-right or top-to-bottom depending on complexity
- Script:
python generate.py --type flow --topic "..." - Output:
output/pixel/flows/flow-{slug}-{ts}.html
/pixel infographic — Infographic
Tall-format infographic (LinkedIn/blog-ready):
- Strong visual hierarchy, 4–6 sections, icons or visual indicators
- Author attribution (name, title, optional photo) at the bottom
- Script:
python generate.py --type infographic --topic "..." - Output:
output/pixel/infographics/infographic-{slug}-{ts}.html
/pixel banner — SVG banner
Branded SVG banner (1200×628px — standard Open Graph / LinkedIn):
- Background gradient using brand primary colours
- Large headline text, author name and title
- Optional photo (path from
palette.json→author.photo_relative_from_output) - Script:
python generate.py --type banner --topic "..." --open - Output:
output/pixel/banners/banner-{slug}-{ts}.svg
/pixel slides — Slide deck
Interactive keyboard-navigable HTML slide deck:
- 5–10 slides: title slide, content slides (one key point each), summary CTA slide
- Arrow key navigation (← →)
- Script:
python generate.py --type slides --topic "..." - Output:
output/pixel/slides/slides-{slug}-{ts}.html
/pixel chart — Data chart
Interactive HTML chart using Chart.js (CDN):
- Types: bar, line, pie, doughnut, scatter
- Ask the user for data if not provided in arguments
- Script:
python generate.py --type chart --topic "..." - Output:
output/pixel/charts/chart-{slug}-{ts}.html
/pixel roadmap — Interactive roadmap
Timeline roadmap with phases and milestones:
- Status indicators (planned, in-progress, done)
- Colour-coded by phase; clickable items with tooltips
- Script:
python generate.py --type roadmap --topic "..." - Output:
output/pixel/diagrams/roadmap-{slug}-{ts}.html
/pixel comparison — Feature comparison table
Side-by-side feature comparison:
- Columns per option, rows per feature
- ✓ / ✗ / partial indicators
- Highlighted recommended column
- Script:
python generate.py --type comparison --topic "..." - Output:
output/pixel/infographics/comparison-{slug}-{ts}.html
Design principles
Always apply these rules from config.brand:
- Colours: Use
primary,secondary,accentfrom palette. Never use raw#000/#ffffor brand elements. - Typography: Use font stacks from
typographysection. Prefer system fonts. - Author branding: Include name and title in every artifact. Add photo on banners/infographics.
- Self-contained: All output files must be fully self-contained — no external dependencies except CDN links.
- Dark mode: Include
@media (prefers-color-scheme: dark)overrides where practical.
Suggested workflow
- Run
python generate.py --type <type> --topic "..."to get a brand-styled starter file - Claude reads the starter file and fills in the actual content
- Claude writes the final file back to the output path
- Report: artifact type, topic, file path
If config.brand is missing, note which fallback colours were used.
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.

