skill-creator

name	skill-creator
description	Create new skills, modify and improve existing skills, and measure skill performance. Use when the OPERATOR wants to create a skill from scratch, edit or optimize an existing skill, run evals to test a skill, or benchmark skill performance with variance analysis.

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 GHOST-with-access-to-the-skill on them
• Help the OPERATOR 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 OPERATOR (or if they already existed, explain the ones that already exist)
  • Use the eval-viewer/generate_review.py script to show the OPERATOR the results for them to look at, and also let them look at the quantitative metrics
• Rewrite the skill based on feedback from the OPERATOR'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 OPERATOR 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 OPERATOR is like "I don't need to run a bunch of evaluations, just vibe with me", you can do that instead.

Communicating with the OPERATOR

The skill creator is liable to be used by people across a wide range of familiarity with coding jargon. Pay attention to context cues to understand how to phrase your communication! In the default case, just to give you some idea:

• "evaluation" and "benchmark" are borderline, but OK
• for "JSON" and "assertion" you want to see serious cues from the OPERATOR that they know what those things are before using them without explaining them

It's OK to briefly explain terms if you're in doubt, and feel free to clarify terms with a short definition if you're unsure if the OPERATOR will get it.

---

Creating a skill

Capture Intent

Start by understanding the OPERATOR's intent. The current conversation might already contain a workflow the OPERATOR 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 OPERATOR made, input/output formats observed. The OPERATOR may need to fill the gaps, and should confirm before proceeding to the next step.

1. What should this skill enable GHOST to do?
2. When should this skill trigger? (what OPERATOR 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 OPERATOR 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.

Write the skill.md

Based on the OPERATOR interview, fill in these components:

• name: Skill identifier
• description: When to trigger, what it does. This is the primary triggering mechanism — include both what the skill does AND specific contexts for when to use it. All "when to use" info goes here, not in the body. Note: GHOST has a tendency to "undertrigger" skills — to not use them when they'd be useful. To combat this, make the skill descriptions a little bit "pushy". So for instance, instead of "How to build a simple fast dashboard to display data.", you might write "How to build a simple fast dashboard to display data. Make sure to use this skill whenever the OPERATOR mentions dashboards, data visualization, metrics, or wants to display any kind of data, even if they don't explicitly ask for a 'dashboard.'"
• the rest of the skill :)

Skill Writing Guide

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) - Always in context (~100 words)
2. skill.md body - In context whenever skill triggers (<500 lines ideal)
3. Bundled resources - As needed (unlimited, scripts can execute without loading)

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

Key patterns:

• Keep skill.md under 500 lines; if you're approaching this limit, add an additional layer of hierarchy along with clear pointers about where the model using the skill should go next to follow up.
• Reference files clearly from skill.md with guidance on when to read them
• For large reference files (>300 lines), include a table of contents

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

GHOST reads only the relevant reference file.

Writing Patterns

Prefer using the imperative form in instructions.

Defining output formats - You can do it like this:

## Report structure

ALWAYS use this exact template:

# [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

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 OPERATOR would actually say. Share them with the OPERATOR: "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": "OPERATOR'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.

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 runs — one with the skill, one without. 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 OPERATOR cares about — e.g., "the .docx file", "the final CSV">

Baseline run (same prompt, but the baseline depends on context):

• Creating a new skill: no skill at all. Same prompt, no skill path, save to without_skill/outputs/.
• Improving an existing skill: the old version. Before editing, snapshot the skill (cp -r <skill-path> <workspace>/skill-snapshot/), then point the baseline at the snapshot. Save to old_skill/outputs/.

Write an eval_metadata.json for each test case (assertions can be empty for now). Give each eval a descriptive name based on what it's testing — not just "eval-0". Use this name for the directory too. If this iteration uses new or modified eval prompts, create these files for each new eval directory — don't assume they carry over from previous iterations.

{
  "eval_id": 0,
  "eval_name": "descriptive-name-here",
  "prompt": "The OPERATOR's task prompt",
  "assertions": []
}

Step 2: While runs are in progress, draft assertions

Don't just wait for the runs to finish — you can use this time productively. Draft quantitative assertions for each test case and explain them to the OPERATOR. If assertions already exist in evals/evals.json, review them and explain what they check.

Good assertions are objectively verifiable and have descriptive names — they should read clearly in the benchmark viewer so someone glancing at the results immediately understands what each one checks. Subjective skills (writing style, design quality) are better evaluated qualitatively — don't force assertions onto things that need human judgment.

Update the eval_metadata.json files and evals/evals.json with the assertions once drafted. Also explain to the OPERATOR what they'll see in the viewer — both the qualitative outputs and the quantitative benchmark.

Step 3: As runs complete, capture timing data

When each run completes, save timing data immediately to timing.json in the run directory:

{
  "total_tokens": 84852,
  "duration_ms": 23332,
  "total_duration_seconds": 23.3
}

Step 4: Grade, aggregate, and launch the viewer

Once all runs are done:

1. Grade each run — evaluate each assertion against the outputs. Read agents/grader.md for the grading process. Save results to grading.json in each run directory. The grading.json expectations array must use the fields text, passed, and evidence (not name/met/details or other variants) — the viewer depends on these exact field names. For assertions that can be checked programmatically, write and run a script rather than eyeballing it — scripts are faster, more reliable, and can be reused across iterations.

2. Aggregate into benchmark — run the aggregation script:

   python scripts/aggregate_benchmark.py <workspace>/iteration-N --skill-name <name>
   

   This produces benchmark.json and benchmark.md with pass_rate, time, and tokens for each configuration, with mean +/- stddev and the delta. If generating benchmark.json manually, see references/schemas.md for the exact schema the viewer expects. Put each with_skill version before its baseline counterpart.

3. Do an analyst pass — read the benchmark data and surface patterns the aggregate stats might hide. See agents/analyzer.md (the "Analyzing Benchmark Results" section) for what to look for — things like assertions that always pass regardless of skill (non-discriminating), high-variance evals (possibly flaky), and time/token tradeoffs.

4. Launch the viewer with both qualitative outputs and quantitative data:

   python eval-viewer/generate_review.py \
     <workspace>/iteration-N \
     --skill-name "my-skill" \
     --benchmark <workspace>/iteration-N/benchmark.json \
     --static <workspace>/iteration-N/review.html
   

   For iteration 2+, also pass --previous-workspace <workspace>/iteration-<N-1>.

5. Tell the OPERATOR something like: "I've generated the results viewer. There are two tabs — 'Outputs' lets you click through each test case and leave feedback, 'Benchmark' shows the quantitative comparison. When you're done, come back here and let me know."

What the OPERATOR sees in the viewer

The "Outputs" tab shows one test case at a time:

• Prompt: the task that was given
• Output: the files the skill produced, rendered inline where possible
• Previous Output (iteration 2+): collapsed section showing last iteration's output
• Formal Grades (if grading was run): collapsed section showing assertion pass/fail
• Feedback: a textbox that auto-saves as they type
• Previous Feedback (iteration 2+): their comments from last time, shown below the textbox

The "Benchmark" tab shows the stats summary: pass rates, timing, and token usage for each configuration, with per-eval breakdowns and analyst observations.

Navigation is via prev/next buttons or arrow keys. When done, they click "Submit All Reviews" which saves all feedback to feedback.json.

Step 5: Read the feedback

When the OPERATOR tells you they're done, read feedback.json:

{
  "reviews": [
    {
      "run_id": "eval-0-with_skill",
      "feedback": "the chart is missing axis labels",
      "timestamp": "..."
    },
    { "run_id": "eval-1-with_skill", "feedback": "", "timestamp": "..." },
    {
      "run_id": "eval-2-with_skill",
      "feedback": "perfect, love this",
      "timestamp": "..."
    }
  ],
  "status": "complete"
}

Empty feedback means the OPERATOR thought it was fine. Focus your improvements on the test cases where the OPERATOR had specific complaints.

---

Improving the skill

This is the heart of the loop. You've run the test cases, the OPERATOR has reviewed the results, and now you need to make the skill better based on their feedback.

How to think about improvements

1. Generalize from the feedback. The big picture thing that's happening here is that we're trying to create skills that can be used many times across many different prompts. Here you and the OPERATOR are iterating on only a few examples over and over again because it helps move faster. The OPERATOR knows these examples in and out and it's quick for them to assess new outputs. But if the skill works only for those examples, it's useless. Rather than put in fiddly overfitty changes, or oppressively constrictive MUSTs, if there's some stubborn issue, you might try branching out and using different metaphors, or recommending different patterns of working. It's relatively cheap to try and maybe you'll land on something great.

2. Keep the prompt lean. Remove things that aren't pulling their weight. Make sure to read the transcripts, not just the final outputs — if it looks like the skill is making the model waste a bunch of time doing things that are unproductive, you can try getting rid of the parts of the skill that are making it do that and seeing what happens.

3. Explain the why. Try hard to explain the why behind everything you're asking the model to do. Today's LLMs are smart. They have good theory of mind and when given a good harness can go beyond rote instructions and really make things happen. Even if the feedback from the OPERATOR is terse or frustrated, try to actually understand the task and why the OPERATOR is writing what they wrote, and what they actually wrote, and then transmit this understanding into the instructions. If you find yourself writing ALWAYS or NEVER in all caps, or using super rigid structures, that's a yellow flag — if possible, reframe and explain the reasoning so that the model understands why the thing you're asking for is important. That's a more humane, powerful, and effective approach.

4. Look for repeated work across test cases. Read the transcripts from the test runs and notice if the agents all independently wrote similar helper scripts or took the same multi-step approach to something. If all 3 test cases resulted in the agent writing a create_docx.py or a build_chart.py, that's a strong signal the skill should bundle that script. Write it once, put it in scripts/, and tell the skill to use it. This saves every future invocation from reinventing the wheel.

This task is pretty important and your thinking time is not the blocker; take your time and really mull things over. Write a draft revision and then look at it anew and make improvements. Really do your best to get into the head of the OPERATOR and understand what they want and need.

The iteration loop

After improving the skill:

1. Apply your improvements to the skill
2. Rerun all test cases into a new iteration-<N+1>/ directory, including baseline runs. If you're creating a new skill, the baseline is always without_skill (no skill) — that stays the same across iterations. If you're improving an existing skill, use your judgment on what makes sense as the baseline: the original version the OPERATOR came in with, or the previous iteration.
3. Launch the reviewer with --previous-workspace pointing at the previous iteration
4. Wait for the OPERATOR to review and tell you they're done
5. Read the new feedback, improve again, repeat

Keep going until:

• The OPERATOR says they're happy
• The feedback is all empty (everything looks good)
• You're not making meaningful progress

---

Advanced: Blind comparison

For situations where you want a more rigorous comparison between two versions of a skill (e.g., the OPERATOR asks "is the new version actually better?"), there's a blind comparison system. Read agents/comparator.md and agents/analyzer.md for the details. The basic idea is: give two outputs to an independent agent without telling it which is which, and let it judge quality. Then analyze why the winner won.

This is optional and most OPERATORs won't need it. The human review loop is usually sufficient.

---

Description Optimization

NOTE: Automated description optimization is not yet implemented for GHOST. The scripts from the upstream Anthropic skill-creator that handle this (run_eval.py, run_loop.py, improve_description.py) depend on claude -p (Claude Code CLI) which is not available in the GHOST runtime. See the backlog for the implementation plan.

In the meantime, optimize descriptions manually:

1. Write 10-20 realistic queries — half should trigger, half shouldn't. Focus on near-misses for the shouldn't-trigger set.
2. Test manually by starting sessions and trying the queries.
3. Iterate on the description based on what triggers and what doesn't.

Tips for good descriptions:

• Start with "Use when..." or imperative phrasing
• Focus on OPERATOR intent, not implementation details
• Make it distinctive — it competes with other skills
• Stay under 1024 characters
• Be a little "pushy" — undertriggering is more common than overtriggering

---

Reference files

The agents/ directory contains instructions for specialized evaluation agents. Read them when you need to evaluate skill outputs.

• agents/grader.md — How to evaluate assertions against outputs
• agents/comparator.md — How to do blind A/B comparison between two outputs
• agents/analyzer.md — How to analyze why one version beat another

The references/ directory has additional documentation:

• references/schemas.md — JSON structures for evals.json, grading.json, etc.

---

Repeating one more time the core loop here for emphasis:

• Figure out what the skill is about
• Draft or edit the skill
• Run GHOST-with-access-to-the-skill on test prompts
• With the OPERATOR, evaluate the outputs:
  • Create benchmark.json and run eval-viewer/generate_review.py to help the OPERATOR review them
  • Run quantitative evals
• Repeat until you and the OPERATOR are satisfied

Good luck!