vt-c-skill-creator

Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.

Plugin: core-standards
Category: Tooling
Command: /vt-c-skill-creator

---

Skill Creator

This skill provides guidance for creating effective skills.

About Skills

Skills are modular, self-contained packages that extend Claude's capabilities by providing specialized knowledge, workflows, and tools. Think of them as "onboarding guides" for specific domains or tasks—they transform Claude from a general-purpose agent into a specialized agent equipped with procedural knowledge that no model can fully possess.

What Skills Provide

1. Specialized workflows - Multi-step procedures for specific domains
2. Tool integrations - Instructions for working with specific file formats or APIs
3. Domain expertise - Company-specific knowledge, schemas, business logic
4. Bundled resources - Scripts, references, and assets for complex and repetitive tasks

Anatomy of a Skill

Every skill consists of a required SKILL.md file and optional bundled resources:

skill-name/
├── SKILL.md (required)
│   ├── YAML frontmatter metadata (required)
│   │   ├── name: (required)
│   │   └── description: (required)
│   └── Markdown instructions (required)
└── Bundled Resources (optional)
    ├── scripts/          - Executable code (Python/Bash/etc.)
    ├── references/       - Documentation intended to be loaded into context as needed
    └── assets/           - Files used in output (templates, icons, fonts, etc.)

SKILL.md (required)

Metadata Quality: The name and description in YAML frontmatter determine when Claude will use the skill. Be specific about what the skill does and when to use it. Use the third-person (e.g. "This skill should be used when..." instead of "Use this skill when...").

Bundled Resources (optional)

Scripts (scripts/)

Executable code (Python/Bash/etc.) for tasks that require deterministic reliability or are repeatedly rewritten.

• When to include: When the same code is being rewritten repeatedly or deterministic reliability is needed
• Example: scripts/rotate_pdf.py for PDF rotation tasks
• Benefits: Token efficient, deterministic, may be executed without loading into context
• Note: Scripts may still need to be read by Claude for patching or environment-specific adjustments

References (references/)

Documentation and reference material intended to be loaded as needed into context to inform Claude's process and thinking.

• When to include: For documentation that Claude should reference while working
• Examples: references/finance.md for financial schemas, references/mnda.md for company NDA template, references/policies.md for company policies, references/api_docs.md for API specifications
• Use cases: Database schemas, API documentation, domain knowledge, company policies, detailed workflow guides
• Benefits: Keeps SKILL.md lean, loaded only when Claude determines it's needed
• Best practice: If files are large (>10k words), include grep search patterns in SKILL.md
• Avoid duplication: Information should live in either SKILL.md or references files, not both. Prefer references files for detailed information unless it's truly core to the skill—this keeps SKILL.md lean while making information discoverable without hogging the context window. Keep only essential procedural instructions and workflow guidance in SKILL.md; move detailed reference material, schemas, and examples to references files.

Assets (assets/)

Files not intended to be loaded into context, but rather used within the output Claude produces.

• When to include: When the skill needs files that will be used in the final output
• Examples: assets/logo.png for brand assets, assets/slides.pptx for PowerPoint templates, assets/frontend-template/ for HTML/React boilerplate, assets/font.ttf for typography
• Use cases: Templates, images, icons, boilerplate code, fonts, sample documents that get copied or modified
• Benefits: Separates output resources from documentation, enables Claude to use files without loading them into context

Progressive Disclosure Design Principle

Skills use a three-level loading system to manage context efficiently:

1. Metadata (name + description) - Always in context (~100 words)
2. SKILL.md body - When skill triggers (<5k words)
3. Bundled resources - As needed by Claude (Unlimited*)

*Unlimited because scripts can be executed without reading into context window.

Cache-friendly authoring: Skill instructions become part of the cached prompt prefix. Place stable content (objective, workflow, criteria) before any dynamic content (dates, counters, session state) to maximize cache hits at 10% of standard token pricing. See create-agent-skills/references/core-principles.md for the full cache-friendly authoring principle with examples.

Skill Creation Process

To create a skill, follow the "Skill Creation Process" in order, skipping steps only if there is a clear reason why they are not applicable.

Step 1: Understanding the Skill with Concrete Examples

Skip this step only when the skill's usage patterns are already clearly understood. It remains valuable even when working with an existing skill.

To create an effective skill, clearly understand concrete examples of how the skill will be used. This understanding can come from either direct user examples or generated examples that are validated with user feedback.

For example, when building an image-editor skill, relevant questions include:

• "What functionality should the image-editor skill support? Editing, rotating, anything else?"
• "Can you give some examples of how this skill would be used?"
• "I can imagine users asking for things like 'Remove the red-eye from this image' or 'Rotate this image'. Are there other ways you imagine this skill being used?"
• "What would a user say that should trigger this skill?"

To avoid overwhelming users, avoid asking too many questions in a single message. Start with the most important questions and follow up as needed for better effectiveness.

Conclude this step when there is a clear sense of the functionality the skill should support.

Step 2: Planning the Reusable Skill Contents

To turn concrete examples into an effective skill, analyze each example by:

1. Considering how to execute on the example from scratch
2. Identifying what scripts, references, and assets would be helpful when executing these workflows repeatedly

Example: When building a pdf-editor skill to handle queries like "Help me rotate this PDF," the analysis shows:

1. Rotating a PDF requires re-writing the same code each time
2. A scripts/rotate_pdf.py script would be helpful to store in the skill

Example: When designing a frontend-webapp-builder skill for queries like "Build me a todo app" or "Build me a dashboard to track my steps," the analysis shows:

1. Writing a frontend webapp requires the same boilerplate HTML/React each time
2. An assets/hello-world/ template containing the boilerplate HTML/React project files would be helpful to store in the skill

Example: When building a big-query skill to handle queries like "How many users have logged in today?" the analysis shows:

1. Querying BigQuery requires re-discovering the table schemas and relationships each time
2. A references/schema.md file documenting the table schemas would be helpful to store in the skill

To establish the skill's contents, analyze each concrete example to create a list of the reusable resources to include: scripts, references, and assets.

Step 3: Initializing the Skill

At this point, it is time to actually create the skill.

Skip this step only if the skill being developed already exists, and iteration or packaging is needed. In this case, continue to the next step.

When creating a new skill from scratch, always run the init_skill.py script. The script conveniently generates a new template skill directory that automatically includes everything a skill requires, making the skill creation process much more efficient and reliable.

Usage:

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

The script:

• Creates the skill directory at the specified path
• Generates a SKILL.md template with proper frontmatter and TODO placeholders
• Creates example resource directories: scripts/, references/, and assets/
• Adds example files in each directory that can be customized or deleted

After initialization, customize or remove the generated SKILL.md and example files as needed.

Step 4: Edit the Skill

When editing the (newly-generated or existing) skill, remember that the skill is being created for another instance of Claude to use. Focus on including information that would be beneficial and non-obvious to Claude. Consider what procedural knowledge, domain-specific details, or reusable assets would help another Claude instance execute these tasks more effectively.

Start with Reusable Skill Contents

To begin implementation, start with the reusable resources identified above: scripts/, references/, and assets/ files. Note that this step may require user input. For example, when implementing a brand-guidelines skill, the user may need to provide brand assets or templates to store in assets/, or documentation to store in references/.

Also, delete any example files and directories not needed for the skill. The initialization script creates example files in scripts/, references/, and assets/ to demonstrate structure, but most skills won't need all of them.

Update SKILL.md

Writing Style: Write the entire skill using imperative/infinitive form (verb-first instructions), not second person. Use objective, instructional language (e.g., "To accomplish X, do Y" rather than "You should do X" or "If you need to do X"). This maintains consistency and clarity for AI consumption.

To complete SKILL.md, answer the following questions:

1. What is the purpose of the skill, in a few sentences?
2. When should the skill be used?
3. In practice, how should Claude use the skill? All reusable skill contents developed above should be referenced so that Claude knows how to use them.

Fork Context (context: fork) Requirements

If the skill uses context: fork, add these mandatory sections:

1. Error Handling (context: fork) — Include the three critical rules: no interactive questions, inline file-write failure reporting, and full formatted output as primary deliverable. See docs/reference/skill-frontmatter.md Fork Context Requirements for the complete rules.

2. TL;DR Summary step — Add a final step that produces a concise summary as the absolute last output. Use the standardized format:

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
TL;DR
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
[Decision result — e.g., PASS/FAIL, GO/NO-GO, or skill-specific outcome]
[Key metrics — e.g., finding counts, files processed, checks passed]
Branch: [branch-name]

Next: [recommended next action]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Constraints: max 10 lines between delimiters, nothing may follow the TL;DR section, and include variants for each possible outcome (success, failure, partial, error).

Step 5: Pressure Test (Strongly Recommended)

Before packaging the skill for distribution, verify it works correctly by testing it against realistic scenarios. This step catches activation failures, missing guidance, and edge cases before the skill reaches production.

Skip this step only for trivial skill updates (typo fixes, minor wording changes) where the skill's core behavior is unchanged.

5a: Generate Test Scenarios

Based on the skill's name, description, and any triggers in its frontmatter, generate 2-3 test scenarios:

1. Happy path — A realistic user request that should activate the skill and produce a good result. Use the primary use case from the skill's description.
2. Edge case — A less obvious request that still falls within the skill's scope. Tests boundary conditions or unusual inputs.
3. Negative case — A request that is superficially similar but should NOT activate this skill. Verifies the skill doesn't trigger on unrelated work.

Each scenario should be a concrete user prompt (1-2 sentences) that another Claude instance could execute. Keep scenarios human-readable and reviewable.

Example scenarios for a pdf-editor skill:

Scenario 1 (happy path): "Rotate this PDF 90 degrees clockwise: /tmp/report.pdf"
Scenario 2 (edge case): "Split this 50-page PDF into individual pages"
Scenario 3 (negative): "Summarize the contents of this PDF for me"

5b: Run Pressure Tests

For each scenario, run two subagent comparisons:

1. Without the skill — Dispatch a general-purpose subagent (model: haiku) with the scenario prompt but WITHOUT loading the skill being tested. Observe:
2. Does it fail or produce inferior results?
3. Does it lack the procedural knowledge the skill provides?

4. For the negative case: Does it handle the request appropriately?

5. With the skill — Dispatch the same subagent type (model: haiku) with the same scenario prompt but WITH the skill loaded. Observe:

6. Does the skill activate and provide the expected guidance?
7. Does the result match the skill's intended behavior?
8. For the negative case: Does the skill correctly NOT interfere?

Use context: fork for all test subagents to ensure isolation (NF-1).

If a context: fork skill is being tested, note in the scenario that the subagent cannot use AskUserQuestion — scenarios must be self-contained (EC5).

5c: Evaluate Results

For each scenario, compare the with-skill and without-skill results:

Pressure Test Results
─────────────────────────────────────────────────────────────────
Scenario 1 (happy path): PASS
  Without skill: Attempted generic approach, missed key workflow steps
  With skill: Followed correct procedure, used bundled script

Scenario 2 (edge case): PASS
  Without skill: Partially succeeded but missed validation step
  With skill: Applied edge case handling from skill instructions

Scenario 3 (negative): PASS
  Without skill: Handled appropriately (unrelated to skill)
  With skill: Skill did not activate (correct behavior)

Overall: 3/3 scenarios passed
─────────────────────────────────────────────────────────────────

Pass criteria: - Happy path: skill must demonstrably improve the result - Edge case: skill must handle the boundary condition - Negative case: skill must NOT interfere with unrelated requests

If a scenario fails: Investigate whether the skill's instructions need improvement (go back to Step 4), or whether the scenario was unrealistic (adjust the scenario).

If a subagent crashes or times out (EC4): Mark the scenario as "inconclusive" and suggest manual testing:

Scenario 2 (edge case): INCONCLUSIVE
  Subagent timed out after 60s. Manual testing recommended.

5d: Record Results

Note the pressure test results in the skill's development context. If the skill passes all scenarios, proceed to Step 6 (Packaging). If any scenario fails, return to Step 4 to improve the skill before packaging.

Step 6: Packaging a Skill

Once the skill is ready, it should be packaged into a distributable zip file that gets shared with the user. The packaging process automatically validates the skill first to ensure it meets all requirements:

scripts/package_skill.py <path/to/skill-folder>

Optional output directory specification:

scripts/package_skill.py <path/to/skill-folder> ./dist

The packaging script will:

1. Validate the skill automatically, checking:
2. YAML frontmatter format and required fields
3. Skill naming conventions and directory structure
4. Description completeness and quality

5. File organization and resource references

6. Package the skill if validation passes, creating a zip file named after the skill (e.g., my-skill.zip) that includes all files and maintains the proper directory structure for distribution.

If validation fails, the script will report the errors and exit without creating a package. Fix any validation errors and run the packaging command again.

Step 7: Iterate

After testing the skill, users may request improvements. Often this happens right after using the skill, with fresh context of how the skill performed.

Iteration workflow: 1. Use the skill on real tasks 2. Notice struggles or inefficiencies 3. Identify how SKILL.md or bundled resources should be updated 4. Implement changes and test again

Architecture Reference

• Compact constraints (for agents): create-agent-skills/references/architecture-constraints.md — actionable rules for building skills and agents
• Full reference (for humans): plugins/core-standards/docs/agent-architecture-patterns.md — conceptual explanations, diagrams, toolkit examples

Agent Isolation Guidance

When creating agents (not skills), determine if the agent needs isolation: worktree in its frontmatter. See create-agent-skills/references/agent-frontmatter-fields.md for the complete frontmatter field reference and the 4-question isolation decision checklist.

Permission Governance

Skills that perform significant work (builds, document generation, multi-file edits) often trigger repeated tool approval prompts, creating "approval fatigue." Two mechanisms exist to pre-authorize tools and reduce interruptions.

allowed-tools for Interactive Skills

For skills that run in the main conversation and need specific tools without approval prompts, declare allowed-tools in the SKILL.md frontmatter:

---
name: vt-c-skill-creator
description: Orchestrates end-to-end markdown documentation pipeline.
---

The listed tools are pre-authorized for the skill session. Claude can use them without triggering approval prompts. Tools not listed still require manual approval.

agent: + context: fork for Autonomous Skills

For skills that should run autonomously in isolation, bind the skill to a pre-authorized agent. The agent's tools: list governs which tools are available in the fork:

---
name: vt-c-skill-creator
description: Execute a complete IMS content development sprint cycle.
context: fork
agent: ims-writer
---

The ims-writer agent (defined in .claude/agents/ or plugins/core-standards/agents/templates/) declares its own tools list, creating an auditable permission boundary for the skill.

Decision Guide

Scenario	Mechanism	Why
Skill runs interactively, needs Read/Write/Edit without prompts	allowed-tools	Pre-authorizes tools in the main conversation
Skill runs autonomously, produces output returned to parent	agent: + context: fork	Isolated execution with defined tool boundary
Skill has minimal tool needs or prompts are acceptable	Neither	Default behavior, no configuration needed
Skill needs some tools pre-authorized but others require approval	allowed-tools (partial list)	Hybrid: listed tools auto-approved, unlisted tools still prompt

Pre-Authorized Agent Templates

The toolkit provides agent templates in plugins/core-standards/agents/templates/ with curated tool sets for common domains:

• ims-writer.md — IMS content authoring (file ops, pandoc, mmdc, git)
• toolkit-developer.md — Plugin and skill development (Edit, Write, Glob, Grep, Bash, TodoWrite)
• docs-pipeline.md — Document generation (file ops, pandoc, mmdc, git)

To bind a skill to a template, set agent: to the template name (without .md). To create a custom permission boundary, copy a template and modify the tools: list.

Permission Discovery

To determine the right tools for a skill:

1. Start restrictive — run the skill with no allowed-tools or a minimal set
2. Observe — note which tool approvals Claude requests during execution
3. Add — add those tools to allowed-tools (or the bound agent's tools: list)
4. Repeat — run the skill again and verify no unnecessary prompts remain

Common permission categories:

Category	Tools
File operations	Read, Write, Edit, MultiEdit, Glob, Grep
Shell execution	Bash
Sub-agent delegation	Task / Agent
Research	WebSearch, WebFetch
Notebook	NotebookEdit, NotebookRead

Safety principle: never grant Bash without considering what commands the skill might run. Prefer narrower tool sets over broad ones. Start from a template in plugins/core-standards/agents/templates/ and narrow down.