omgkit 2.3.0 → 2.4.0

package/templates/stdrules/SKILL_STANDARDS.md

@@ -0,0 +1,490 @@
+# OMGKIT Skill Standards v2.0
+
+> Aligned with Claude Platform Official Best Practices
+> Source: https://platform.claude.com/docs/en/agents-and-tools/agent-skills/best-practices
+
+This document defines the quality standards for all OMGKIT skills based on Claude's official skill authoring guidelines.
+
+---
+
+## Core Principles
+
+### 1. Concise is Key
+
+The context window is a public good. Your Skill shares it with:
+- System prompt
+- Conversation history
+- Other Skills' metadata
+- User's actual request
+
+**Default assumption**: Claude is already very smart. Only add context Claude doesn't already have.
+
+Challenge each piece of information:
+- "Does Claude really need this explanation?"
+- "Can I assume Claude knows this?"
+- "Does this paragraph justify its token cost?"
+
+### 2. Progressive Disclosure
+
+SKILL.md serves as an overview that points Claude to detailed materials as needed:
+- Keep SKILL.md body **under 500 lines**
+- Split content into separate files when approaching this limit
+- Claude reads files only when needed
+
+### 3. Degrees of Freedom
+
+Match specificity to task fragility:
+
+| Freedom Level | When to Use | Example |
+|---------------|-------------|---------|
+| **High** (text instructions) | Multiple valid approaches | Code review guidelines |
+| **Medium** (pseudocode/params) | Preferred pattern exists | Report templates |
+| **Low** (specific scripts) | Operations are fragile | Database migrations |
+
+---
+
+## Frontmatter Requirements
+
+```yaml
+---
+name: skill-name
+description: Third-person description of what it does and when to use it
+---
+```
+
+### Name Rules
+- Maximum **64 characters**
+- Lowercase letters, numbers, hyphens only
+- No XML tags, no reserved words ("anthropic", "claude")
+- **Recommended**: Use gerund form (verb + -ing)
+
+**Good names** (gerund form):
+- `processing-pdfs`
+- `analyzing-data`
+- `managing-databases`
+- `testing-code`
+
+**Acceptable alternatives**:
+- Noun phrases: `pdf-processing`, `data-analysis`
+- Action-oriented: `process-pdfs`, `analyze-data`
+
+**Avoid**:
+- Vague: `helper`, `utils`, `tools`
+- Generic: `documents`, `data`, `files`
+
+### Description Rules
+- Maximum **1024 characters**
+- **Must be third person** (not "I can" or "You can use")
+- Include BOTH what it does AND when to use it
+- Be specific, include key terms
+
+**Good descriptions**:
+```yaml
+# Good - specific and includes trigger context
+description: Extracts text and tables from PDF files, fills forms, merges documents. Use when working with PDF files or when the user mentions PDFs, forms, or document extraction.
+
+# Good - includes specific triggers
+description: Generates descriptive commit messages by analyzing git diffs. Use when the user asks for help writing commit messages or reviewing staged changes.
+```
+
+**Bad descriptions**:
+```yaml
+description: Helps with documents  # Too vague
+description: Processes data  # No trigger context
+description: I can help you with PDFs  # Wrong person
+```
+
+---
+
+## Skill Structure
+
+### Minimal SKILL.md (Recommended for Most Skills)
+
+```markdown
+---
+name: processing-pdfs
+description: Extracts text and tables from PDF files, fills forms, merges documents. Use when working with PDF files or when the user mentions PDFs, forms, or document extraction.
+---
+
+# Processing PDFs
+
+## Quick Start
+
+Extract text with pdfplumber:
+```python
+import pdfplumber
+with pdfplumber.open("file.pdf") as pdf:
+    text = pdf.pages[0].extract_text()
+```
+
+## Features
+
+- **Text extraction**: Use pdfplumber for text
+- **Table extraction**: See [TABLES.md](TABLES.md)
+- **Form filling**: See [FORMS.md](FORMS.md)
+- **Merging**: See [MERGE.md](MERGE.md)
+
+## Workflows
+
+### Standard Extraction
+1. Load PDF with pdfplumber
+2. Extract text from each page
+3. Clean and format output
+
+For advanced workflows, see [WORKFLOWS.md](WORKFLOWS.md).
+```
+
+### Directory Structure for Complex Skills
+
+```
+skill-name/
+├── SKILL.md              # Main instructions (<500 lines)
+├── REFERENCE.md          # API reference (loaded as needed)
+├── EXAMPLES.md           # Usage examples (loaded as needed)
+├── WORKFLOWS.md          # Complex workflows (loaded as needed)
+└── scripts/
+    ├── utility.py        # Utility script (executed, not loaded)
+    └── validate.py       # Validation script
+```
+
+---
+
+## Content Guidelines
+
+### What to Include
+
+1. **Quick start** - Minimal code to get started
+2. **Feature overview** - Brief descriptions with links to details
+3. **Common workflows** - Step-by-step for typical tasks
+4. **References to detailed files** - For advanced content
+
+### What NOT to Include
+
+1. **Explanations of basic concepts** - Claude knows what PDFs are
+2. **Multiple library options** - Pick one default, mention alternatives only if necessary
+3. **Verbose descriptions** - Be direct and concise
+4. **Time-sensitive information** - Use "old patterns" section if needed
+
+### Example: Concise vs Verbose
+
+**Good (≈50 tokens)**:
+```markdown
+## Extract PDF Text
+
+Use pdfplumber for text extraction:
+```python
+import pdfplumber
+with pdfplumber.open("file.pdf") as pdf:
+    text = pdf.pages[0].extract_text()
+```
+```
+
+**Bad (≈150 tokens)**:
+```markdown
+## Extract PDF Text
+
+PDF (Portable Document Format) files are a common file format that contains
+text, images, and other content. To extract text from a PDF, you'll need to
+use a library. There are many libraries available for PDF processing, but we
+recommend pdfplumber because it's easy to use and handles most cases well.
+First, you'll need to install it using pip. Then you can use the code below...
+```
+
+---
+
+## Progressive Disclosure Patterns
+
+### Pattern 1: High-Level Guide with References
+
+```markdown
+---
+name: processing-documents
+description: Processes various document formats including PDF, DOCX, and XLSX. Use when working with documents or file conversion.
+---
+
+# Document Processing
+
+## Quick Reference
+
+| Format | Library | Guide |
+|--------|---------|-------|
+| PDF | pdfplumber | [PDF.md](PDF.md) |
+| DOCX | python-docx | [DOCX.md](DOCX.md) |
+| XLSX | openpyxl | [XLSX.md](XLSX.md) |
+
+## Common Operations
+
+**Extract text**: See format-specific guide above
+**Convert formats**: See [CONVERT.md](CONVERT.md)
+**Batch processing**: See [BATCH.md](BATCH.md)
+```
+
+### Pattern 2: Domain-Specific Organization
+
+```
+bigquery-skill/
+├── SKILL.md
+└── reference/
+    ├── finance.md
+    ├── sales.md
+    └── product.md
+```
+
+```markdown
+# BigQuery Analysis
+
+## Datasets
+
+| Domain | Schema | Guide |
+|--------|--------|-------|
+| Finance | revenue, billing | [reference/finance.md](reference/finance.md) |
+| Sales | opportunities, pipeline | [reference/sales.md](reference/sales.md) |
+| Product | API usage, features | [reference/product.md](reference/product.md) |
+```
+
+### Pattern 3: Conditional Details
+
+```markdown
+# DOCX Processing
+
+## Creating Documents
+Use docx-js for new documents. See [DOCX-JS.md](DOCX-JS.md).
+
+## Editing Documents
+For simple edits, modify XML directly.
+
+**For tracked changes**: See [REDLINING.md](REDLINING.md)
+**For OOXML details**: See [OOXML.md](OOXML.md)
+```
+
+---
+
+## Workflows and Feedback Loops
+
+### Use Checklists for Complex Tasks
+
+```markdown
+## Form Processing Workflow
+
+Copy this checklist and track progress:
+
+```
+Progress:
+- [ ] Step 1: Analyze form (run analyze_form.py)
+- [ ] Step 2: Create field mapping (edit fields.json)
+- [ ] Step 3: Validate mapping (run validate_fields.py)
+- [ ] Step 4: Fill form (run fill_form.py)
+- [ ] Step 5: Verify output (run verify_output.py)
+```
+
+**Step 1**: Run `python scripts/analyze_form.py input.pdf`
+**Step 2**: Edit `fields.json` with values
+**Step 3**: Run `python scripts/validate_fields.py fields.json`
+**Step 4**: Run `python scripts/fill_form.py input.pdf fields.json output.pdf`
+**Step 5**: Run `python scripts/verify_output.py output.pdf`
+
+If verification fails, return to Step 2.
+```
+
+### Implement Validation Loops
+
+```markdown
+## Edit Process
+
+1. Make edits to document
+2. **Validate immediately**: `python validate.py`
+3. If validation fails:
+   - Review error message
+   - Fix issues
+   - Run validation again
+4. **Only proceed when validation passes**
+5. Save output
+```
+
+---
+
+## Reference Files (for Detailed Content)
+
+Keep references **one level deep** from SKILL.md:
+
+**Good** (one level):
+```
+SKILL.md → reference.md → (no further links)
+SKILL.md → examples.md → (no further links)
+```
+
+**Bad** (too deep):
+```
+SKILL.md → advanced.md → details.md → (lost information)
+```
+
+### Structure Long Reference Files
+
+For files >100 lines, include table of contents:
+
+```markdown
+# API Reference
+
+## Contents
+- Authentication
+- Core Methods
+- Advanced Features
+- Error Handling
+- Examples
+
+## Authentication
+...
+```
+
+---
+
+## Anti-Patterns to Avoid
+
+### ❌ Avoid Windows Paths
+```markdown
+# Bad
+scripts\helper.py
+
+# Good
+scripts/helper.py
+```
+
+### ❌ Avoid Too Many Options
+```markdown
+# Bad
+"You can use pypdf, or pdfplumber, or PyMuPDF, or pdf2image..."
+
+# Good
+"Use pdfplumber for text extraction.
+For scanned PDFs requiring OCR, use pdf2image with pytesseract instead."
+```
+
+### ❌ Avoid Time-Sensitive Info
+```markdown
+# Bad
+"If before August 2025, use old API..."
+
+# Good
+## Current Method
+Use v2 API endpoint.
+
+## Legacy (Deprecated)
+<details>
+<summary>v1 API (deprecated 2025-08)</summary>
+Old endpoint details...
+</details>
+```
+
+### ❌ Avoid Inconsistent Terminology
+```markdown
+# Bad
+Mix of "API endpoint", "URL", "route", "path"
+
+# Good
+Always use "API endpoint" consistently
+```
+
+---
+
+## Skill Length Guidelines
+
+| Skill Type | SKILL.md Max | Use Reference Files When |
+|------------|--------------|--------------------------|
+| Simple | 100 lines | Rarely needed |
+| Standard | 200-300 lines | 1-2 reference files |
+| Complex | 400-500 lines | 3+ reference files |
+
+**Hard limit**: SKILL.md body should never exceed 500 lines.
+
+---
+
+## Testing Checklist
+
+Before finalizing a skill:
+
+### Core Quality
+- [ ] Name uses gerund form (or acceptable alternative)
+- [ ] Name is ≤64 characters, lowercase, hyphens only
+- [ ] Description is third person
+- [ ] Description includes what it does AND when to use it
+- [ ] SKILL.md body is under 500 lines
+- [ ] Only essential context included (Claude already knows basics)
+- [ ] References are one level deep
+- [ ] No time-sensitive information
+- [ ] Consistent terminology throughout
+
+### Structure
+- [ ] Quick start section with minimal code
+- [ ] Feature overview with links to details
+- [ ] Workflows have clear steps
+- [ ] Complex content in separate reference files
+
+### If Includes Scripts
+- [ ] Scripts handle errors explicitly
+- [ ] No "voodoo constants" (all values justified)
+- [ ] Clear whether to execute or read scripts
+- [ ] Forward slashes in all paths
+
+---
+
+## Migration from v1 Skills
+
+To convert existing verbose skills:
+
+1. **Extract quick start**: Keep only the minimal getting-started code
+2. **Create reference files**: Move detailed content to REFERENCE.md, EXAMPLES.md, etc.
+3. **Simplify descriptions**: Remove explanations Claude already knows
+4. **Add navigation**: Replace inline content with links to reference files
+5. **Verify length**: Ensure SKILL.md is under 500 lines
+
+### Example Migration
+
+**Before (v1 - 800 lines)**:
+```markdown
+# MongoDB
+
+MongoDB is a document database... [200 lines of explanation]
+
+## Schema Design
+[300 lines of patterns and examples]
+
+## Query Optimization
+[200 lines of techniques]
+
+## Best Practices
+[100 lines of do's and don'ts]
+```
+
+**After (v2 - 150 lines)**:
+```markdown
+# MongoDB
+
+## Quick Start
+
+```python
+from pymongo import MongoClient
+client = MongoClient("mongodb://localhost:27017/")
+db = client["mydb"]
+```
+
+## Features
+
+| Feature | Guide |
+|---------|-------|
+| Schema Design | [SCHEMA.md](SCHEMA.md) |
+| Query Optimization | [QUERIES.md](QUERIES.md) |
+| Indexing | [INDEXES.md](INDEXES.md) |
+| Transactions | [TRANSACTIONS.md](TRANSACTIONS.md) |
+
+## Common Patterns
+
+- **CRUD operations**: See Quick Start above
+- **Aggregation**: See [AGGREGATION.md](AGGREGATION.md)
+- **Replication**: See [REPLICATION.md](REPLICATION.md)
+```
+
+---
+
+*Version: 2.0*
+*Based on: Claude Platform Best Practices (2025)*
+*Last Updated: 2024-12-31*