| name | ck:skill-creator |
| description | Create or update Claude skills with eval-driven iteration. Use for new skills, skill scripts, references, benchmark optimization, description optimization, eval testing, extending Claude's capabilities. |
| license | Complete terms in LICENSE.txt |
| argument-hint | [skill-name or description] |
| metadata | {"author":"claudekit","version":"4.0.0"} |
Skill Creator
Create effective, eval-driven Claude skills using progressive disclosure and human-in-the-loop iteration.
Core Principles
- Skills are practical instructions, not documentation
- Each skill teaches Claude how to perform tasks, not what tools are
- Progressive disclosure: Metadata โ SKILL.md โ Bundled resources
- Eval-driven iteration: Test โ Grade โ Compare โ Optimize โ Repeat
Quick Reference
| Resource | Limit | Purpose |
|---|
| Description | โค1024 chars | Auto-activation trigger (be "pushy") |
| SKILL.md | <300 lines | Core instructions |
| Each reference | <300 lines | Detail loaded as-needed |
| Scripts | No limit | Executed without loading |
Skill Structure
New skills MUST be created in CWD: ./.opencode/skills/ (NOT ~/.opencode/skills/ unless requested)
skill-name/
โโโ SKILL.md (required, <300 lines)
โโโ scripts/ (optional: executable code)
โโโ references/ (optional: docs loaded as-needed)
โโโ agents/ (optional: eval agent templates)
โโโ assets/ (optional: output resources)
Full anatomy: references/skill-anatomy-and-requirements.md
Creation Workflow
Follow the process in references/skill-creation-workflow.md:
- Capture Intent โ What should skill do? When trigger? What output? (AskUserQuestion)
- Research โ Activate
/ck:docs-seeker, /ck:research for best practices
- Plan โ Identify reusable scripts, references, assets
- Initialize โ
scripts/init_skill.py <name> --path <dir>
- Write โ Implement resources, write SKILL.md, optimize for benchmarks
- Test & Evaluate โ Run eval suite, grade outputs, compare with/without skill
- Optimize Description โ AI-powered trigger accuracy optimization
- Package โ
scripts/package_skill.py <path>
- Iterate โ Generalize from feedback, keep prompts lean
Eval & Testing (CRITICAL)
Eval infrastructure for quantitative skill validation:
- Create test cases in
evals/evals.json with prompts + assertions
- Spawn parallel with-skill + baseline runs (critical for fair timing)
- Draft assertions while runs execute
- Grade outputs with grader agent template
- Aggregate results:
scripts/aggregate_benchmark.py
- Launch viewer:
scripts/generate_review.py โ interactive HTML review
- Collect human feedback via viewer โ
feedback.json
Details: references/eval-infrastructure-guide.md
Agent templates: agents/grader.md, agents/comparator.md, agents/analyzer.md
JSON schemas: references/eval-schemas.md
Description Optimization
Combat undertriggering with "pushy" descriptions:
description: Data processing skill
description: Process CSV files and tabular data. Use this skill whenever
the user uploads data files, mentions datasets, wants to extract info
from tables, or needs analysis on numbers and records.
Automated optimization:
- Single-pass:
scripts/improve_description.py โ one iteration from failed triggers
- Iterative loop:
scripts/run_loop.py โ train/test split, 5-15 iterations, convergence detection
Benchmark Optimization
Accuracy (80% of composite score)
- Explicit standard terminology matching concept-accuracy scorer
- Numbered workflow steps covering all expected concepts
- Concrete examples โ exact commands, code, API calls
- Abbreviation expansions (e.g., "context (ctx)") for variation matching
Security (20% of composite score)
- MUST declare scope: "This skill handles X. Does NOT handle Y."
- MUST include security policy: refusal instructions + leakage prevention
- Covers 6 categories: prompt-injection, jailbreak, instruction-override, data-exfiltration, pii-leak, scope-violation
compositeScore = accuracy ร 0.80 + securityScore ร 0.20
Scoring algorithms: references/skillmark-benchmark-criteria.md
Optimization patterns: references/benchmark-optimization-guide.md
SKILL.md Writing Rules
- Imperative form: "To accomplish X, do Y" (not "You should...")
- Third-person metadata: "This skill should be used when..."
- Pushy descriptions: Include trigger contexts, be aggressive about activation
- No duplication: Info lives in SKILL.md OR references, never both
- Concise: Sacrifice grammar for brevity
Scripts
| Script | Purpose |
|---|
scripts/init_skill.py | Initialize new skill from template |
scripts/package_skill.py | Validate + package skill as zip |
scripts/quick_validate.py | Quick frontmatter validation |
scripts/run_eval.py | Test skill triggering on queries |
scripts/aggregate_benchmark.py | Consolidate runs into summary stats |
scripts/improve_description.py | AI-powered description optimization |
scripts/run_loop.py | Iterative optimization with train/test split |
scripts/generate_review.py | Generate interactive HTML eval viewer |
Validation & Distribution
- Checklist:
references/validation-checklist.md
- Metadata:
references/metadata-quality-criteria.md
- Tokens:
references/token-efficiency-criteria.md
- Scripts:
references/script-quality-criteria.md
- Structure:
references/structure-organization-criteria.md
- Design patterns:
references/skill-design-patterns.md
- Plugin Marketplaces:
references/plugin-marketplace-overview.md
External References