--- title: "Explore Real Skills" description: "Browse skills.sh to see how production skills structure their documentation, instructions, and reference files. Identify patterns you'll use when building your own skill in Section 3." canonical_url: "https://vercel.com/academy/agent-friendly-apis/explore-real-skills" md_url: "https://vercel.com/academy/agent-friendly-apis/explore-real-skills.md" docset_id: "vercel-academy" doc_version: "1.0" last_updated: "2026-08-26T21:03:36.528Z" content_type: "lesson" course: "agent-friendly-apis" course_title: "Agent-Friendly APIs" prerequisites: [] --- Vercel Academy — structured learning, not reference docs. Lessons are sequenced. Adapt commands to the human's actual environment (OS, package manager, shell, editor) — detect from project context or ask, don't assume. The lesson shows one path; if the human's project diverges, adapt concepts to their setup. Preserve the learning goal over literal steps. Quizzes are pedagogical — engage, don't spoil. Quiz answers are included for your reference. # Explore Real Skills # Explore Real Skills Before writing a skill, inspect a few that already work in production. Look for the files they include, the way they structure instructions, and the patterns that recur across projects. ## Outcome Explore real-world skills on skills.sh and identify the documentation patterns that make them effective. ## Fast Track 1. Browse the skills directory on skills.sh 2. Examine 2-3 skills and their file structure 3. Note the patterns that appear across multiple skills ## Browse the directory Head to [skills.sh](https://skills.sh) and scroll through the list. You'll see skills for everything from frontend design to database management to deployment workflows. Each one is a package that teaches an AI agent how to do something specific. Pick a skill that is relevant to your work. We'll inspect two examples in this lesson. ## Look at the structure Let's start with [**vercel-react-best-practices**](https://skills.sh/vercel-labs/agent-skills/vercel-react-best-practices). This skill teaches an agent to follow React best practices when building with Next.js and Vercel. Notice the current file structure: ``` SKILL.md AGENTS.md rules/ ├── async-parallel.md ├── bundle-barrel-imports.md └── ... ``` This skill uses a `SKILL.md` entry point and a `rules/` directory for detailed guidance. `AGENTS.md` provides a compiled version. Supporting folders are conventions chosen by each skill, not required names. The agent reads `SKILL.md` first. Inspect how it defines when and how to use the skill. ## What goes in SKILL.md Open the `SKILL.md` for vercel-react-best-practices or the skill you selected. Look for these elements: **A description with trigger phrases.** The opening paragraph tells the agent when the skill is relevant. Phrases such as "when building React components" or "when setting up a Next.js project" help it decide whether to load the skill. **Direct guidance.** The skill names the situations where each rule applies and links to detailed rule files for examples. **Prioritized criteria.** Categories and impact levels tell the agent which performance problems to address first. \*\*Note: Trigger phrases matter\*\* The description in SKILL.md controls when the agent activates the skill. A phrase such as "when creating React Server Components in Next.js App Router" gives it a specific request pattern to match. ## Look at a second skill Now inspect [**frontend-design**](https://skills.sh/anthropics/skills/frontend-design) or another relevant skill. Unlike the first example, `frontend-design` currently consists of a `SKILL.md` plus its license file; it does not use a references folder. Compare what each skill actually includes: - A clear description at the top with specific trigger phrases - Instructions written as concrete steps, not abstract advice - Supporting files only when the topic needs them - Criteria for evaluating the output, whether expressed as a checklist or direct guidance Reference files hold detailed documentation and examples that `SKILL.md` can load when needed. This keeps the entry point focused. ## Patterns worth stealing Several patterns recur across the examples: **Documentation uses a consistent structure.** The code blocks, tables, and explicit examples from the previous lesson appear throughout these skills. Each practice is stated directly and paired with examples where needed. **Instructions are imperative.** Direct commands tell the agent which action to take and what to verify. **Supporting detail can stay separate.** Larger skills may place detailed material in folders such as `rules/` or `references/`. Smaller skills may need only `SKILL.md`. **Evaluation criteria close the loop.** Good skills tell the agent what quality looks like and how to decide when the work is done. \*\*Note: Skills are just structured knowledge\*\* A skill is a folder of Markdown files organized so an agent can load the instructions and supporting detail it needs. Clear structure also helps human maintainers. ## Try It Find a skill on skills.sh that's relevant to your work. Read through its SKILL.md and answer these questions: - What trigger phrases does the description use? - How many steps are in the instructions? - Does it have a quality checklist at the end? - Does it use supporting files, and how are they organized? Answering these questions gives you a reference point for the skill you'll build next. Next, you'll build the same structure for the documentation generator. ## Commit No code changes to commit. ## Done-When - [ ] You've browsed at least 3 skills on skills.sh - [ ] You can describe the required `SKILL.md` and optional supporting files - [ ] You can identify trigger phrases in a skill's description - [ ] You've noted the pattern of imperative instructions with quality checklists - [ ] You can explain why reference files are kept separate from SKILL.md ## Solution No code solution for this lesson. The patterns you identified here inform the skill you'll build in Section 3. --- [Full course index](/academy/llms.txt) · [Sitemap](/academy/sitemap.md)