Ai discoverable portfolio
Skill obadadallo95/ai-native-portfolio-skills/skills/ai-discoverable-portfolio
An opinionated skill system for AI-native builders who want portfolios with real identity, proof, taste, and AI discoverability.
npx -y skills add obadadallo95/ai-native-portfolio-skills --skill ai-discoverable-portfolioAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Helps builders create portfolios that are understandable to AI agents, LLM-based discovery tools, summarization systems, and other machine readers without weakening the human experience.
SKILL.md
3.3 KB, 635 tokens by cl100k_base, as published. Nobody here has run it
AI-Discoverable Portfolio
This skill exists because a portfolio is no longer only for human readers.
It is now also read by:
- AI agents
- LLM-based discovery tools
- summarization systems
- crawlers
- systems that extract who a builder is, what they do, and whether they are relevant
The goal is not to make a portfolio robotic.
The goal is to make it:
- readable to humans
- interpretable to models
- structured for extraction
- easier to discover and summarize correctly
Core Rule
Do not treat AI readability as an afterthought.
Treat it as a second layer that supports discoverability while preserving authored presentation for humans.
What AI Systems Need
AI systems generally perform better when a portfolio clearly expresses:
- who the builder is
- what kind of work they do
- who they work with
- what projects matter most
- what technologies or domains recur
- what differentiates their approach
- how to contact or engage them
If those facts are buried only inside visual atmosphere, the model may miss them.
Structural Guidelines
Favor:
- explicit project titles
- clear role descriptions
- concise summaries
- stable section headings
- structured metadata where it helps
- dedicated machine-readable files when appropriate
- consistent naming across pages
Avoid:
- relying only on vague slogans
- hiding critical meaning inside decorative UI
- making every section poetic but hard to summarize
- inconsistent project naming
- pages that look rich visually but are weak semantically
Human + AI Dual-Layer Rule
A strong portfolio should have two layers:
Human layer
- emotion
- voice
- design
- pacing
- visual authority
AI-readable layer
- explicit summaries
- machine-readable files
- structured metadata
- semantic headings
- discoverable project framing
Neither layer should destroy the other.
Implementation Directions
Depending on the stack, consider:
- strong semantic HTML
- stable heading hierarchy
- project summaries near each project
- metadata pages or text mirrors
llms.txt,ai.txt, or similar discovery-oriented files- consistent portfolio-wide terminology
- clear structured descriptions for projects, roles, and outcomes
Audit Questions
Before shipping, ask:
- Can a model quickly tell who this builder is?
- Can a model distinguish the top projects from secondary work?
- Can a model extract the builder's type of work without guessing?
- Can a model understand the difference between identity copy and proof copy?
- Are key claims supported by explicit project descriptions?
- Are important pages and artifacts discoverable outside of the visual UI?
Failure Mode
This skill fails when:
- the portfolio is beautiful but hard to summarize
- the builder's identity is emotionally strong but semantically fuzzy
- projects are visually impressive but structurally unclear
- AI systems can crawl the site but not interpret it well
Success Condition
This skill succeeds when the portfolio can be:
- experienced with taste by humans
- understood accurately by AI systems
- surfaced more clearly in AI-mediated discovery