Anatomy of a Skill

Every skill generated by SkillThis follows the same two-part structure: YAML frontmatter (metadata) and a markdown body (content). This page walks through a complete example.

Full Example

Here’s a complete skill generated from a technical recruiter’s input:

---
name: recruiting-engineering-talent
description: Screens and evaluates engineering candidates using structured frameworks. Use when sourcing candidates, screening resumes, conducting technical interviews, or managing hiring pipelines for engineering roles.
---

# Recruiting Engineering Talent

## Quick Start

Given a job requirement, produce a qualified candidate shortlist:

1. Parse the job req for must-have vs nice-to-have skills
2. Build boolean search strings for LinkedIn Recruiter
3. Source 40-50 candidates, prioritizing active GitHub contributors
4. Screen using MATCH framework (below)
5. Present top 5 candidates with match rationale

## Workflow

### Sourcing
1. Translate job requirements into LinkedIn boolean queries
2. Search GitHub for active contributors in relevant technologies
3. Check Stack Overflow profiles for domain expertise
4. Target 40-50 initial candidates per role

### Screening (MATCH Framework)
For each candidate, evaluate:
- **M**otivation: Why are they looking? What excites them?
- **A**ptitude: Can they grow into the role?
- **T**echnical depth: Do they have the core skills?
- **C**ulture fit: Do they align with team values?
- **H**unger: Are they driven and proactive?

Score each dimension 1-5. Minimum total score: 18/25 to advance.

### Interview Prep
- Send candidates a prep doc 48 hours before each round
- Include: company context, interviewer bios, what to expect
- Debrief with hiring manager within 24 hours of each round

## Examples

**Example 1: Senior Backend Engineer**
Input: "Need a senior backend engineer, Go experience, distributed systems"
Output:
- Boolean search: `("senior" OR "staff") AND ("Go" OR "Golang") AND "distributed systems"`
- Sourced 45 candidates from LinkedIn + GitHub
- MATCH-screened to 12, presented 5
- 3 advanced to onsite, 1 offer accepted (6-week timeline)

**Example 2: Early Career Frontend**
Input: "Junior React developer, design sensibility preferred"
Output:
- Search focused on bootcamp grads and recent CS grads with portfolio sites
- Weighted MATCH toward Hunger and Aptitude over Technical depth
- Sourced 60, screened 20, presented 8
- 4 advanced, 2 offers (one accepted)

## Best Practices

- Always prep candidates before interviews (reduces no-shows by 40%)
- Debrief hiring managers same-day while impressions are fresh
- Track all interactions in ATS (Greenhouse, Lever, etc.)
- Use the "soft close" approach: help candidates decide, never hard sell
- Build relationships even with rejected candidates (future pipeline)

## Common Pitfalls

- Screening for keywords instead of capability
- Skipping the motivation check (leads to early attrition)
- Not calibrating with the hiring manager on what "senior" means
- Sending generic outreach messages (response rate drops 60%)

The Two Parts

Part 1: YAML Frontmatter

The frontmatter sits between --- markers at the top. It contains metadata that AI assistants use to decide when to invoke the skill.

---
name: recruiting-engineering-talent
description: Screens and evaluates engineering candidates...
---

See YAML Frontmatter for the full reference.

Part 2: Markdown Body

The body contains the actual knowledge. It’s structured in sections that follow a consistent pattern.

See Skill Sections for what each section does.

How AI Assistants Read Skills

When you paste a skill into an AI assistant:

The description tells the AI when this skill is relevant
The Quick Start gives immediate context for what to do
The Workflow provides the step-by-step process to follow
The Examples show what good output looks like
The Best Practices and Pitfalls set guardrails

The AI uses all of this context to respond in a way that follows your methodology rather than giving generic advice.