skill-creator

name	Skill Creator
description	Standards for creating, testing, and optimizing Agent Skills. Use when creating, improving, catching regressions, measuring trigger rates, or writing eval cases for any skill.
metadata	{"labels":["meta","standard","instruction-design","token-efficient","eval","testing"],"triggers":{"files":["SKILL.md","metadata.json","evals/evals.json"],"keywords":["create skill","new standard","writing rules","high density","test skill","eval skill","trigger rate","optimize","description","skill regression","improve skill"]}}

Agent Skill Creator Standard

Priority: P0 (CRITICAL)

Strict guidelines for High-Density Agent Skills. Maximize info/token ratio.

Core Principles (Token Economy First ⚡)

• Progressive Loading: Load only essential content initially.
• Lazy References: Move detailed examples to references/.
• Imperative Compression: Use verbs, abbreviations, bullet points.
• Context Limits: Cursor(~100k), Claude(~200k), Windsurf(~32k).

Three-Level Loading System

1. Metadata: Triggers → AGENTS.md index (Proactive Activation)
2. SKILL.md: Body < 100 lines → Core guidelines (When triggered)
3. Resources: references/, scripts/, assets/ → Deep knowledge (On-demand)

Writing Rules

• Imperative: Start with verbs. "Use BLoC" not "You should use BLoC".
• Token Economy: Skip articles. Use standard abbreviations. Bullets > paragraphs.
• Structure:
  1. Mandatory Frontmatter (YAML: name, description, metadata labels & triggers).
  2. Priority: P0 (Critical), P1 (Standard), P2 (Optional).
  3. Guidelines: Imperative Do's.
  4. Anti-Patterns: Strict format Don'ts.
  5. References: Links to lazy-loaded files.

Strict Size Limits

Element	Limit	Action if Exceeded
SKILL.md total	100 lines	Extract to references/
Inline code block	10 lines	Extract to references/
Anti-pattern item	15 words	Compress to imperative
Tables	8 rows	Extract to references/

Strict Formatting Rules

• Anti-Patterns: **No X**: Do Y[, not Z]. [Context <= 15 words]
  • Example: **No Logic in Builder**: Perform calculations in BLoC, not UI.
• No Redundancy: Do not repeat frontmatter descriptions.
• Oversized Skills: If SKILL.md >100 lines, extract step-by-step guides and complex scenarios to references/.
• Nested Formatting: Avoid **Bold**: \More Bold``.

Test, Measure & Iterate

After writing a skill draft, validate it before shipping:

1. Write eval cases: Create evals/evals.json with 2–3 realistic prompts (see testing.md).
2. Design trigger queries: Generate 8–10 should-trigger and 8–10 should-not-trigger queries to measure description accuracy.
3. Optimize description: Make it "pushy" — list explicit trigger contexts, not just what it does.
4. Catch regressions: Snapshot the skill before editing; compare before/after outputs for existing test cases.
5. Iterate: Re-run evals after each change; stop when trigger rate ≥ 80% on held-out queries.

See Testing, Trigger Rate & Regression Guide for eval schema, query design rules, and regression protocol.

Resources & Deep Knowledge

• Resource Organization & Directory Structure
• Skill Creation Lifecycle
• Anti-Patterns Details
• Creation Template
• Testing, Trigger Rate & Regression Guide