Skill Architect: The Authoritative Meta-Skill

The unified authority for creating expert-level Agent Skills. Encodes the knowledge that separates a skill that merely exists from one that activates precisely, teaches efficiently, and makes users productive immediately.

Philosophy

Great skills are progressive disclosure machines. They encode real domain expertise (shibboleths), not surface instructions. They follow a three-layer architecture: lightweight metadata for discovery, lean SKILL.md for core process, and reference files for deep dives loaded only on demand.

When to Use This Skill

• Creating new skills from scratch or from existing expertise
• Auditing/reviewing skills for quality, activation, and progressive disclosure
• Improving activation rates and reducing false positives
• Encoding domain expertise (shibboleths, anti-patterns, temporal knowledge)
• Designing skills that subagents consume effectively
• Building self-contained tools (scripts, MCPs, subagents)
• Debugging why skills don't activate or activate incorrectly

NOT for: General Claude Code features (slash commands, MCP server implementation), non-skill coding advice or code review, debugging runtime errors (use domain-specific skills), template generation without real domain expertise to encode.

Quick Wins (Immediate Improvements)

For existing skills, apply in priority order:

1. Tighten description -- Follow

   [What] [When to use]. NOT for [Exclusions]

   formula
2. Check line count -- SKILL.md must be <500 lines; move depth to

   /references

3. Add NOT clause -- Prevent false activation with explicit exclusions
4. Add 1-2 anti-patterns -- Use shibboleth template (Novice/Expert/Timeline)
5. Remove dead files -- Delete unreferenced files in

   scripts/

   and

   references/

   (no phantoms)
6. Test activation -- Write 5 queries that should trigger and 5 that shouldn't

Progressive Disclosure Architecture

Skills use three-layer loading. The runtime scans metadata at startup, loads SKILL.md on activation, and pulls reference files only when the agent decides it needs them.

name

description

Critical rules:

• Keep SKILL.md under 500 lines. Move depth to

  /references

  .
• Reference files are NOT auto-loaded. Only SKILL.md enters context on activation.
• In SKILL.md, list each reference file with a 1-line description of when to consult it.
• Never instruct "read all reference files before starting." Instead: "Read only the files relevant to the current step."

Frontmatter Rules

Required Fields

name

description

[What] [When to use]. NOT for [Exclusions]

Optional Fields

allowed-tools

Read,Write,Grep

argument-hint

"[path] [format]"

license

MIT

disable-model-invocation

true

/skill-name

true

user-invocable

true

context

fork

fork

agent

context: fork

code-reviewer

model

sonnet

hooks

metadata

author: your-org

Custom Keys (Safe to Use)

Custom keys like

category

tags

version

metadata:

Invalid Keys (Confusingly Similar to Valid Ones)

# These look like valid keys but aren't -- use the correct alternatives
tools: Read,Write           # Use 'allowed-tools' instead
integrates_with: [...]      # Use SKILL.md body text instead
outputs: [...]              # Use SKILL.md Output Format section instead
dependencies: [...]         # Use SKILL.md body text (not a real frontmatter key)
bundled-resources: [...]    # Use SKILL.md body text (not a real frontmatter key)

Common mistakes that prevent loading:

tools:

allowed-tools:

[Read, Write]

python scripts/validate_skill.py <path>

Platform constraints (name: 64 chars, description: 1024 chars, upload: 8MB, 8 skills/request max, no XML tags): See

references/claude-extension-taxonomy.md

Description Formula

Pattern:

[What it does] [When to use -- be slightly pushy]. NOT for [Exclusions].

The description is the most important line for activation. Claude's runtime scans descriptions to decide which skill to load. Claude evaluates descriptions semantically, not via keyword matching. It reasons about whether your description covers the user's intent. Claude also tends to undertrigger -- make descriptions slightly pushy to combat this.

Full guide with more examples: See

references/description-guide.md

SKILL.md Template

---
name: your-skill-name
description: [What it does] [When to use -- be slightly pushy]. NOT for [Exclusions].
allowed-tools: Read,Write
---

# Skill Name
[One sentence purpose]

## When to Use
Use for: [A, B, C with specific trigger keywords]
NOT for: [D, E, F -- explicit boundaries]

## Core Process
[Mermaid diagrams for decisions/flows. See visual-artifacts.md for catalog]

## Anti-Patterns
### [Pattern Name]
**Novice**: [Wrong assumption]
**Expert**: [Why it's wrong + correct approach]
**Timeline**: [When this changed, if temporal]

## References
- `references/guide.md` -- Consult when [specific situation]
- `references/examples.md` -- Consult for [worked examples of X]

The 6-Step Skill Creation Process

flowchart LR
  S1[1. Gather Examples] --> S2[2. Plan Contents]
  S2 --> S3[3. Initialize]
  S3 --> S4[4. Write Skill]
  S4 --> S5[5. Validate]
  S5 --> S6{Errors?}
  S6 -->|Yes| S4
  S6 -->|No| S7[6. Ship & Iterate]

Step 1: Gather Concrete Examples

Collect 3-5 real queries that should trigger this skill, and 3-5 that should NOT.

Step 2: Plan Reusable Contents

Identify scripts, references, assets that prevent re-work. Also identify shibboleths: domain algorithms, temporal knowledge, framework evolution, common pitfalls.

Step 3: Initialize

scripts/init_skill.py <skill-name> --path <output-directory>

Step 4: Write the Skill

Order: Scripts first (working code) -> References next (domain knowledge) -> SKILL.md last (core process, reference index).

Write in imperative form: "To accomplish X, do Y" not "You should do X."

Step 5: Validate

python scripts/validate_skill.py <path>
python scripts/check_self_contained.py <path>

Fix ERRORS then WARNINGS then SUGGESTIONS.

Step 6: Iterate

After real-world use: notice struggles, improve SKILL.md and resources, update CHANGELOG.md.

Designing Skills for Subagent Consumption

Three Skill-Loading Layers

1. Preloaded (2-5 core skills): Injected into the subagent's system context.
2. Dynamically selected: Subagent receives a catalog (name + 1-line description) and picks 1-3 matching skills.
3. Execution-time: Subagent reads each skill's "When to use" section, follows numbered steps, respects output contracts, runs QA checks.

Subagent Prompt Structure

Four sections: Identity (narrow role), Skill usage rules (skills as standard operating procedures), Task loop (restate, select skills, plan, execute, validate, return), Constraints (quality bar, safety, tie-breaking).

Full templates and orchestration patterns: See

references/subagent-design.md

Visual Artifacts: Mermaid Diagrams

For humans, diagrams render as visual flowcharts, state machines, and timelines. For agents, Mermaid is a text-based graph DSL --

A -->|Yes| B

Rule: If a skill describes a process, decision tree, architecture, state machine, or data relationship, include a Mermaid diagram.

flowchart TD

sequenceDiagram

stateDiagram-v2

timeline

erDiagram

mindmap

quadrantChart

architecture-beta

Full catalog (all 23 types) with syntax, examples, and YAML config: See

references/visual-artifacts.md

Encoding Shibboleths

Expert knowledge that separates novices from experts. Things LLMs get wrong due to outdated training data or cargo-culted patterns.

Shibboleth Template

### Anti-Pattern: [Name]
**Novice**: "[Wrong assumption]"
**Expert**: [Why it's wrong, with evidence]
**Timeline**: [Date]: [Old way] -> [Date]: [New way]
**LLM mistake**: [Why LLMs suggest the old pattern]
**Detection**: [How to spot this in code/config]

Full catalog with case studies: See

references/antipatterns.md

Self-Contained Tools and the Extension Taxonomy

Skills are one of seven Claude extension types. Most skills should include scripts. MCPs are only for auth/state boundaries. Plugins are for sharing skills across teams/community.

user-invocable: true

Full taxonomy with examples and common mistakes: See

references/claude-extension-taxonomy.md

references/self-contained-tools.md

references/plugin-architecture.md

Tool Permissions (Least Privilege)

allowed-tools

Read,Grep,Glob

Read,Write,Edit

Read,Write,Bash(npm:*,git:*)

Bash

Anti-Pattern Summary

/references

[What] [When to use]. NOT for [Exclusions]

Full case studies: See

references/antipatterns.md

Validation Checklist

[ ] SKILL.md exists and is <500 lines
[ ] Frontmatter has name + description (minimum required)
[ ] Description follows [What][When to use] NOT [Exclusions] formula
[ ] Description is specific and context-rich (semantic activation, not keyword lists)
[ ] Name and description are aligned (not contradictory)
[ ] At least 1 anti-pattern with shibboleth template
[ ] All referenced files actually exist (no phantoms)
[ ] Scripts work (not templates), have clear CLI, handle errors
[ ] Reference files each have a 1-line purpose in SKILL.md
[ ] Processes/decisions/lifecycles use Mermaid diagrams, not prose
[ ] CHANGELOG.md tracks version history
[ ] If subagent-consumed: output contracts are defined
[ ] Skill passes its own validation tools (meta-consistency)

Run automated checks:

python scripts/validate_skill.py <path>

python scripts/validate_mermaid.py <path>

Success Metrics

Reference Files

Consult these for deep dives -- they are NOT loaded by default:

references/knowledge-engineering.md

references/description-guide.md

references/antipatterns.md

references/self-contained-tools.md

references/subagent-design.md

references/claude-extension-taxonomy.md

references/plugin-architecture.md

references/visual-artifacts.md

references/mcp-template.md

references/subagent-template.md

references/scoring-rubric.md

references/skill-composition.md

references/skill-lifecycle.md

references/activation-debugging.md

agents/cross-evaluator.md