laguagu/claude-code-nextjs-skills/skill-creator

Skill Creator

A skill for creating new skills and iteratively improving them.

At a high level, the process of creating a skill goes like this:

• Decide what you want the skill to do and roughly how it should do it
• Write a draft of the skill
• Create a few test prompts and run claude-with-access-to-the-skill on them
• Help the user evaluate the results both qualitatively and quantitatively
• While the runs happen in the background, draft some quantitative evals if there aren't any (if there are some, you can either use as is or modify if you feel something needs to change about them). Then explain them to the user (or if they already existed, explain the ones that already exist)
• Use the eval-viewer/generate_review.py script to show the user the results for them to look at, and also let them look at the quantitative metrics
• Rewrite the skill based on feedback from the user's evaluation of the results (and also if there are any glaring flaws that become apparent from the quantitative benchmarks)
• Repeat until you're satisfied
• Expand the test set and try again at larger scale

Your job when using this skill is to figure out where the user is in this process and then jump in and help them progress through these stages. So for instance, maybe they're like "I want to make a skill for X". You can help narrow down what they mean, write a draft, write the test cases, figure out how they want to evaluate, run all the prompts, and repeat.

On the other hand, maybe they already have a draft of the skill. In this case you can go straight to the eval/iterate part of the loop.

Of course, you should always be flexible and if the user is like "I don't need to run a bunch of evaluations, just vibe with me", you can do that instead.

Then after the skill is done (but again, the order is flexible), you can also run the skill description improver, which we have a whole separate script for, to optimize the triggering of the skill.

Adapt terminology to the user; explain evaluation criteria when needed.

Creating a skill

Capture Intent

Start by understanding the user's intent. The current conversation might already contain a workflow the user wants to capture (e.g., they say "turn this into a skill"). If so, extract answers from the conversation history first — the tools used, the sequence of steps, corrections the user made, input/output formats observed. The user may need to fill the gaps, and should confirm before proceeding to the next step.

1. What should this skill enable Claude to do?
2. When should this skill trigger? (what user phrases/contexts)
3. What's the expected output format?
4. Should we set up test cases to verify the skill works? Skills with objectively verifiable outputs (file transforms, data extraction, code generation, fixed workflow steps) benefit from test cases. Skills with subjective outputs (writing style, art) often don't need them. Suggest the appropriate default based on the skill type, but let the user decide.

Interview and Research

Proactively ask questions about edge cases, input/output formats, example files, success criteria, and dependencies. Wait to write test prompts until you've got this part ironed out.

Check available MCPs - if useful for research (searching docs, finding similar skills, looking up best practices), research in parallel via subagents if available, otherwise inline. Come prepared with context to reduce burden on the user.

Write the SKILL.md

Based on the user interview, fill in these components:

• name: Skill identifier (1–64 chars, lowercase a-z/0-9/hyphens, must match directory name)
• description: What it does and when to use it (1–1024 chars, third person). Name concrete use contexts and distinguish adjacent skills. Adjust its breadth from observed trigger misses and false positives rather than assuming every description should trigger more often.
• argument-hint (optional, Claude Code extension — not part of the agentskills.io spec; other-client support varies and strict validators flag it): Shown in the skill list to guide users (e.g., "[file or directory]")
• compatibility (optional): Platform/environment requirements, 1–500 chars
• the rest of the skill :)

Skill Writing Guide

Authoring Best Practices

Reference: https://platform.claude.com/docs/en/agents-and-tools/agent-skills/best-practices — the full checklist lives in references/best-practices.md; read it before writing or reviewing skill content. The four principles that matter most while drafting:

• Claude is already very smart. Add what the agent lacks: local decisions and their reasons, measured findings, non-obvious contracts and source routes that work where the skill runs (a URL is not reachable in every client). Keep a short resource map or exact schema when it prevents a real mistake; avoid duplicating the codebase or teaching routine discovery.
• Concise is key. Cut repetition and basics while preserving the task's useful knowledge and finish line. Prefer an exact name or short snippet when it clarifies a non-obvious API contract or required output; ordinary APIs do not need an example each.
• Match freedom to fragility. Text guidelines where many approaches work; exact scripts where consistency is critical. Most skills land in between.
• Descriptions in third person. "Processes Excel files", not "I can help you process Excel files" — the description is injected into the system prompt, and mixed point-of-view hurts discovery.

Anatomy of a Skill

skill-name/
├── SKILL.md (required)
│   ├── YAML frontmatter (name, description required)
│   └── Markdown instructions
└── Bundled Resources (optional)
    ├── scripts/    - Executable code for deterministic/repetitive tasks
    ├── references/ - Docs loaded into context as needed
    └── assets/     - Files used in output (templates, icons, fonts)

Progressive Disclosure

Skills use a three-level loading system:

1. Metadata (name + description) - Client discovery/selection; listing behavior varies
2. SKILL.md body - In context whenever skill triggers (<500 lines ideal)
3. Bundled resources - As needed; nothing loads until read, and scripts can run without loading their source. Bundled files also work where the agent has no web access

These word counts are approximate and you can feel free to go longer if needed.

Key patterns:

• Aim for a concise core; 500 lines is a recommendation, not a format limit. Link separate topics with when-to-read guidance, after removing duplication.
• Reference files clearly from SKILL.md with guidance on when to read them
• Add a compact table of contents to long, multi-topic references when it helps locate the relevant section; a line count alone does not require one.

Domain organization: When a skill supports multiple domains/frameworks, organize by variant:

cloud-deploy/
├── SKILL.md (workflow + selection)
└── references/
    ├── aws.md
    ├── gcp.md
    └── azure.md

Claude reads only the relevant reference file.

Principle of Lack of Surprise

This goes without saying, but skills must not contain malware, exploit code, or any content that could compromise system security. A skill's contents should not surprise the user in their intent if described. Don't go along with requests to create misleading skills or skills designed to facilitate unauthorized access, data exfiltration, or other malicious activities. Things like a "roleplay as an XYZ" are OK though.

Writing Patterns

Prefer using the imperative form in instructions.

Defining output formats - Specify an exact template when a downstream workflow or parser requires it. For open-ended work, name the information needed and use the project's format. For a fixed report contract:

## Report structure
Use this structure when producing the report required by this workflow:
# [Title]
## Executive summary
## Key findings
## Recommendations

Examples pattern - It's useful to include examples. You can format them like this (but if "Input" and "Output" are in the examples you might want to deviate a little):

## Commit message format
**Example 1:**
Input: Added user authentication with JWT tokens
Output: feat(auth): implement JWT-based authentication

Gotchas Section

Include non-obvious pitfalls when they prevent a plausible mistake: environment quirks, implicit assumptions or behavior the agent cannot infer from the code. Do not add a Gotchas section merely to repeat the main instructions.

Pre-publish Checklist

Before declaring a skill done, verify the format and referenced files: name matches the directory (lowercase+hyphens, 1–64 chars), and description is specific, third person and at most 1024 characters. Assess size, reference structure and trigger behavior against the task. Preserve necessary version bounds or dated local observations with a current source; avoid volatile claims without one.

Writing Style

Try to explain to the model why things are important in lieu of heavy-handed musty MUSTs. Use theory of mind and try to make the skill general and not super-narrow to specific examples. Start by writing a draft and then look at it with fresh eyes and improve it.

Test Cases

After writing the skill draft, come up with 2-3 realistic test prompts — the kind of thing a real user would actually say. Share them with the user: [you don't have to use this exact language] "Here are a few test cases I'd like to try. Do these look right, or do you want to add more?" Then run them.

Save test cases to evals/evals.json. Don't write assertions yet — just the prompts. You'll draft assertions in the next step while the runs are in progress.

{
  "skill_name": "example-skill",
  "evals": [
    {
      "id": 1,
      "prompt": "User's task prompt",
      "expected_output": "Description of expected result",
      "files": []
    }
  ]
}

See references/schemas.md for the full schema (including the assertions field, which you'll add later).

Running and evaluating test cases

This section is one continuous sequence — don't stop partway through. Do NOT use /skill-test or any other testing skill.

Put results in <skill-name>-workspace/ as a sibling to the skill directory. Within the workspace, organize results by iteration (iteration-1/, iteration-2/, etc.) and within that, each test case gets a directory (eval-0/, eval-1/, etc.). Don't create all of this upfront — just create directories as you go.

Step 1: Spawn all runs (with-skill AND baseline) in the same turn

For each test case, spawn two subagents in the same turn — one with the skill, one without. This is important: don't spawn the with-skill runs first and then come back for baselines later. Launch everything at once so it all finishes around the same time.

With-skill run:

Execute this task:
- Skill path: <path-to-skill>
- Task: <eval prompt>
- Input files: <eval files if any, or "none">
- Save outputs to: <workspace>/iteration-<N>/eval-<ID>/with_skill/outputs/
- Outputs to save: <what the user cares about — e.g., "the .docx file", "the final CSV">