danielmiessler
Create and validate skills. USE WHEN create skill, new skill, skill structure, canonicalize. SkillSearch('createskill') for docs.
bunx add-skill danielmiessler/personal_ai_infrastructure -s CreateSkillLoadingβ¦
Before executing, check for user customizations at:
~/.claude/skills/CORE/USER/SKILLCUSTOMIZATIONS/CreateSkill/
If this directory exists, load and apply any PREFERENCES.md, configurations, or resources found there. These override default behavior. If the directory does not exist, proceed with skill defaults.
You MUST send this notification BEFORE doing anything else when this skill is invoked.
Send voice notification:
curl -s -X POST http://localhost:8888/notify \
-H "Content-Type: application/json" \
-d '{"message": "Running the WORKFLOWNAME workflow in the CreateSkill skill to ACTION"}' \
> /dev/null 2>&1 &
Output text notification:
Running the **WorkflowName** workflow in the **CreateSkill** skill to ACTION...
This is not optional. Execute this curl command immediately upon skill invocation.
MANDATORY skill creation framework for ALL skill creation requests.
Before creating ANY skill, READ:
~/.claude/skills/CORE/SkillSystem.mdCanonical example to follow: ~/.claude/skills/_BLOGGING/SKILL.md
All naming must use TitleCase (PascalCase).
| Component | Format | Example |
|---|---|---|
| Skill directory | TitleCase | Blogging, Daemon, CreateSkill |
| Workflow files | TitleCase.md | Create.md, UpdateDaemonInfo.md |
| Reference docs | TitleCase.md | ProsodyGuide.md, ApiReference.md |
| Tool files | TitleCase.ts | ManageServer.ts |
| Help files | TitleCase.help.md | ManageServer.help.md |
Wrong (NEVER use):
createskill, create-skill, CREATE_SKILLcreate.md, update-info.md, SYNC_REPO.mdCRITICAL: Keep folder structure FLAT - maximum 2 levels deep.
Maximum depth: skills/SkillName/Category/
skills/SkillName/SKILL.md # Skill root
skills/SkillName/Workflows/Create.md # Workflow - one level deep - GOOD
skills/SkillName/Tools/Manage.ts # Tool - one level deep - GOOD
skills/SkillName/QuickStartGuide.md # Context file - in root - GOOD
skills/SkillName/Examples.md # Context file - in root - GOOD
skills/SkillName/Resources/Guide.md # Context files go in root, NOT Resources/
skills/SkillName/Docs/Examples.md # Context files go in root, NOT Docs/
skills/SkillName/Workflows/Category/File.md # THREE levels - NO
skills/SkillName/Templates/Primitives/File.md # THREE levels - NO
skills/SkillName/Tools/Utils/Helper.ts # THREE levels - NO
ONLY these subdirectories are allowed:
Context files (documentation, guides, references) go in the skill ROOT, NOT in subdirectories.
If you need to organize many workflows, use clear filenames instead of subdirectories:
Workflows/CompanyDueDiligence.mdWorkflows/Company/DueDiligence.mdSee: ~/.claude/skills/CORE/SkillSystem.md (Flat Folder Structure section)
For skills with SKILL.md > 100 lines: Use dynamic loading to reduce context on skill invocation.
Session startup: Only frontmatter loads for routing Skill invocation: Full SKILL.md loads Context files: Load only when workflows reference them
SKILL.md = Minimal (30-50 lines) - loads on skill invocation
Additional .md files = Context files - SOPs for specific aspects (loaded on-demand)
NEVER create Context/ or Docs/ subdirectories.
Additional .md files ARE the context files. They live directly in skill root.
WRONG:
skills/Art/
βββ SKILL.md
βββ Context/ β NEVER CREATE THIS
βββ Aesthetic.md
CORRECT:
skills/Art/
βββ SKILL.md
βββ Aesthetic.md β
Context file in skill root
βββ Examples.md β
Context file in skill root
βββ Tools.md β
Context file in skill root
The skill directory IS the context.
skills/Art/
βββ SKILL.md # 40 lines - minimal routing
βββ Aesthetic.md # Context file - SOP for aesthetic
βββ Examples.md # Context file - SOP for examples
βββ Tools.md # Context file - SOP for tools
βββ Workflows/ # Workflows
β βββ Essay.md
βββ Tools/ # CLI tools
βββ Generate.ts
---
name: SkillName
description: Brief. USE WHEN triggers.
---
# SkillName
Brief description.
## Workflow Routing
| Trigger | Workflow |
|---------|----------|
| "trigger" | `Workflows/WorkflowName.md` |
## Quick Reference
**Key points** (3-5 bullet points)
**Full Documentation:**
- Detail 1: `SkillSearch('skillname detail1')` β loads Detail1.md
- Detail 2: `SkillSearch('skillname detail2')` β loads Detail2.md
β Use dynamic loading for:
β Don't use for:
See: ~/.claude/skills/CORE/SkillSystem.md (Dynamic Loading Pattern section)
| Workflow | Trigger | File |
|---|---|---|
| CreateSkill | "create a new skill" | Workflows/CreateSkill.md |
| ValidateSkill | "validate skill", "check skill" | Workflows/ValidateSkill.md |
| UpdateSkill | "update skill", "add workflow" | Workflows/UpdateSkill.md |
| CanonicalizeSkill | "canonicalize", "fix skill structure" | Workflows/CanonicalizeSkill.md |
Example 1: Create a new skill from scratch
User: "Create a skill for managing my recipes"
β Invokes CreateSkill workflow
β Reads SkillSystem.md for structure requirements
β Creates skill directory with TitleCase naming
β Creates SKILL.md, Workflows/, tools/
β Generates USE WHEN triggers based on intent
Example 2: Fix an existing skill that's not routing properly
User: "The research skill isn't triggering - validate it"
β Invokes ValidateSkill workflow
β Checks SKILL.md against canonical format
β Verifies TitleCase naming throughout
β Verifies USE WHEN triggers are intent-based
β Reports compliance issues with fixes
Example 3: Canonicalize a skill with old naming
User: "Canonicalize the daemon skill"
β Invokes CanonicalizeSkill workflow
β Renames workflow files to TitleCase
β Updates routing table to match
β Ensures Examples section exists
β Verifies all checklist items
Use when you need to run Flow type checking, or when seeing Flow type errors in React code.
Use when you want to validate changes before committing, or when you need to check all React contribution requirements.
Use when feature flag tests fail, flags need updating, understanding @gate pragmas, debugging channel-specific test failures, or adding new flags to React.
Use when you need to check feature flag states, compare channels, or debug why a feature behaves differently across release channels.