Km updates

.claude/skills/km-review.md

@@ -0,0 +1,332 @@
+# KM Documentation Review
+
+Applies comprehensive Knowledge Management (KM) team standards to README and documentation files, providing structured feedback and optional auto-fixes.
+
+## Usage
+
+```
+/km-review <file-path>
+/km-review <file-path> --fix
+/km-review <file-path> --score-only
+/km-review <file-path> --ai-only (skip docs-linter, AI review only)
+```
+
+## Description
+
+This skill performs a comprehensive KM standards review on documentation files, checking for:
+- Heading capitalization (Chicago title case for all headings H1–H6)
+- Punctuation: Oxford comma in lists of three or more items; em dashes (—) without surrounding spaces for parenthetical phrases; en dashes (–) for ranges; no spaced em dashes ( — )
+- List formatting consistency (dashes vs asterisks)
+- Link phrasing ("see" vs "refer to", "about" vs "around")
+- Code block formatting (no $ prompts, proper language tags)
+- SAP terminology consistency (on-premise, SAP BTP, etc.)
+- Document structure and completeness
+- Technical accuracy and clarity
+
+## Instructions
+
+When this skill is invoked:
+
+1. **Parse the command arguments**:
+   - Extract file path (required)
+   - Check for `--fix` flag (optional: apply safe fixes automatically)
+   - Check for `--score-only` flag (optional: return only quality score)
+   - Check for `--ai-only` flag (optional: skip docs-linter, use AI review only)
+
+2. **Validate the file**:
+   - Verify the file exists
+   - Confirm it's a markdown file (.md extension)
+   - Read the file content
+
+3. **Load KM standards context**:
+   - Read `prompts/km-doc-review.md` for the complete review prompt
+   - Read `docs/km-style-guide.md` for standards reference
+   - Read `training-data/km-feedback-patterns.json` if available
+
+4. **Optional: Run docs-linter first** (unless --ai-only flag is provided):
+   - Run: `node docs-linter/src/cli.js validate [FILE] --json`
+   - If successful, collect linter findings for merging with AI review
+   - If linter fails, continue with AI-only review
+   - Note: docs-linter is now fully ESM-compatible and operational
+
+5. **Spawn a Task agent** with `subagent_type: "general-purpose"` and pass this prompt:
+
+```
+You are a senior technical documentation reviewer for the SAP Knowledge Management (KM) team. Your task is to review the following markdown file against established KM documentation standards.
+
+## File to Review
+
+File path: [FILE_PATH]
+
+Content:
+```markdown
+[FILE_CONTENT]
+```
+
+## KM Standards to Apply
+
+[INSERT FULL CONTENT FROM prompts/km-doc-review.md HERE]
+
+## Additional Context from KM Style Guide
+
+[INSERT KEY SECTIONS FROM docs/km-style-guide.md HERE]
+
+## Docs-Linter Findings (if available)
+
+[IF docs-linter was run successfully, INSERT ITS FINDINGS HERE with this format:]
+
+The automated docs-linter has already identified the following objective issues:
+
+```json
+[LINTER_FINDINGS_JSON]
+```
+
+Your task is to:
+1. Validate and enhance these linter findings with context
+2. Identify additional issues the linter may have missed
+3. Provide comprehensive explanations for why each issue matters
+4. Add any subjective/contextual issues that require human judgment
+
+## Review Instructions
+
+1. **Analyze the document** against all KM standards
+2. **Categorize findings** by:
+   - Structural issues (heading hierarchy, section ordering, TOC)
+   - Formatting issues (lists, code blocks, links)
+   - Content issues (clarity, terminology, completeness)
+   - Technical issues (accuracy, examples, working links)
+
+3. **For each finding, provide**:
+   - Severity: critical | major | minor | info
+   - Category: structural | formatting | content | technical
+   - Line number (if applicable)
+   - Current text (exact quote)
+   - Suggested fix (exact replacement)
+   - Explanation (why this matters per KM standards)
+   - Is it safe to auto-fix? (boolean)
+
+4. **Calculate a quality score** (0-100):
+   - Start at 100
+   - Deduct points based on severity and frequency
+   - Use the scoring rules from the KM standards
+
+5. **Provide recommendations**:
+   - Prioritized list of improvements
+   - References to KM style guide sections
+   - Examples from quality documentation
+
+## Output Format
+
+Return your review as structured JSON:
+
+```json
+{
+  "file": "[FILE_PATH]",
+  "qualityScore": {
+    "overall": 85,
+    "breakdown": {
+      "structure": 9,
+      "clarity": 8,
+      "maintainability": 9,
+      "developerUsability": 8
+    },
+    "rationale": "Brief explanation of score"
+  },
+  "findings": [
+    {
+      "severity": "major",
+      "category": "formatting",
+      "line": 42,
+      "current": "### Cloud Connector Configuration",
+      "suggested": "### Cloud Connector Configuration",
+      "explanation": "All headings (H1–H6) must use Chicago title case per KM standards. 'Configuration' is a major word and should be capitalized.",
+      "safeToAutoFix": true
+    }
+  ],
+  "summary": {
+    "total": 12,
+    "critical": 0,
+    "major": 4,
+    "minor": 6,
+    "info": 2,
+    "autoFixable": 8
+  },
+  "recommendations": [
+    "Review heading capitalization rules in KM style guide",
+    "Consider updating Table of Contents to use dashes instead of asterisks"
+  ]
+}
+```
+```
+
+5. **Process the agent's response**:
+   - Parse the JSON output
+   - If docs-linter was used, merge and deduplicate findings
+   - If `--score-only` flag: Display only the quality score and exit
+   - Otherwise: Display formatted findings
+
+6. **If `--fix` flag is provided**:
+   - Filter findings where `safeToAutoFix: true`
+   - Apply fixes to the file using the Edit tool
+   - Create a backup comment showing what was changed
+   - Display summary of applied fixes
+
+7. **Format the output** for the user:
+
+```
+📊 KM Documentation Review: [FILE_NAME]
+
+Quality Score: [SCORE]/100
+├─ Structure: [SCORE]/10
+├─ Clarity: [SCORE]/10
+├─ Maintainability: [SCORE]/10
+└─ Developer Usability: [SCORE]/10
+
+Findings: [TOTAL] ([CRITICAL] critical, [MAJOR] major, [MINOR] minor, [INFO] info)
+
+[If --fix applied:]
+✅ Auto-fixed [COUNT] issues
+
+Critical Issues:
+[List critical findings with line numbers and explanations]
+
+Major Issues:
+[List major findings]
+
+Minor Issues:
+[List minor findings]
+
+Recommendations:
+- [Recommendation 1]
+- [Recommendation 2]
+
+[If not --fix:]
+To automatically fix safe issues, run: /km-review [FILE_PATH] --fix
+```
+
+## Examples
+
+### Example 1: Basic Review
+
+**Input:**
+```
+/km-review misc/onpremise/README.md
+```
+
+**Output:**
+```
+📊 KM Documentation Review: misc/onpremise/README.md
+
+Quality Score: 88/100
+├─ Structure: 9/10
+├─ Clarity: 9/10
+├─ Maintainability: 9/10
+└─ Developer Usability: 8/10
+
+Findings: 5 (0 critical, 2 major, 3 minor, 0 info)
+
+Major Issues:
+
+Line 86: Heading capitalization
+  Current: ### Cloud Connector Configuration
+  Suggested: ### Cloud Connector Configuration
+  → All headings (H1–H6) use Chicago title case — this heading is already correct
+
+Line 90: Heading capitalization
+  Current: ### SAP BTP destination
+  Suggested: ### SAP BTP Destination
+  → All headings must use Chicago title case; 'Destination' is a major word and must be capitalized
+
+Minor Issues:
+
+Line 7: Table of Contents formatting
+  Current: * [Overview](#overview)
+  Suggested: - [Overview](#overview)
+  → Use dashes (-) instead of asterisks (*) for list markers
+
+[... more findings ...]
+
+Recommendations:
+- Review heading capitalization rules in docs/km-style-guide.md
+- Update Table of Contents to use consistent dash markers
+- All findings are safe to auto-fix
+
+To automatically fix these issues, run: /km-review misc/onpremise/README.md --fix
+```
+
+### Example 2: With Auto-Fix
+
+**Input:**
+```
+/km-review misc/destinations/README.md --fix
+```
+
+**Output:**
+```
+📊 KM Documentation Review: misc/destinations/README.md
+
+Quality Score: 92/100 (before fixes)
+
+Findings: 8 (0 critical, 2 major, 6 minor, 0 info)
+
+✅ Auto-fixed 8 issues:
+  ✓ Fixed 2 heading capitalization issues
+  ✓ Fixed 4 "refer to" → "see" replacements
+  ✓ Fixed 1 "around" → "about" replacement
+  ✓ Fixed 1 Table of Contents format
+
+Changes applied to misc/destinations/README.md
+
+New Quality Score: 98/100
+└─ All auto-fixable issues resolved
+
+Remaining Issues: None
+
+The file has been updated with KM standards. Review the changes with:
+  git diff misc/destinations/README.md
+```
+
+### Example 3: Score Only
+
+**Input:**
+```
+/km-review misc/sslcerts/README.md --score-only
+```
+
+**Output:**
+```
+📊 KM Quality Score: misc/sslcerts/README.md
+
+Overall: 95/100
+├─ Structure: 10/10
+├─ Clarity: 9/10
+├─ Maintainability: 10/10
+└─ Developer Usability: 9/10
+
+Status: ✅ Excellent - Meets KM standards
+
+For detailed findings, run: /km-review misc/sslcerts/README.md
+```
+
+## Error Handling
+
+- If file doesn't exist: "Error: File not found: [FILE_PATH]"
+- If not a markdown file: "Error: Only markdown files (.md) are supported"
+- If agent fails: "Error: KM review agent failed. Please try again or review manually."
+- If --fix fails: "Error: Failed to apply fixes. File unchanged. Findings available above."
+
+## Implementation Notes
+
+1. **Performance**: The review agent may take 10-20 seconds for large files
+2. **Safety**: Only apply auto-fixes where `safeToAutoFix: true`
+3. **Context**: The agent has full access to KM standards and training data
+4. **Scoring**: Quality scores are calibrated against repository quality examples
+5. **Git-Friendly**: Applied fixes generate clean diffs suitable for commits
+
+## Related Files
+
+- `prompts/km-doc-review.md` - Full KM review prompt (v2.0)
+- `docs/km-style-guide.md` - Complete KM standards reference
+- `docs/km-pr-checklist.md` - PR review checklist
+- `training-data/km-feedback-patterns.json` - Feedback pattern analysis
+- `.claude/skills/customer-tone.md` - Related skill for customer communications

.github/PULL_REQUEST_TEMPLATE.md

@@ -8,3 +8,22 @@
 
 - [ ] Sample tested and runs without errors
 - [ ] No sensitive data included
+- [ ] Documentation reviewed and validated (if README changes included)
+
+## Documentation Validation
+
+If your PR includes changes to README.md files, you can validate them locally before submitting:
+
+```bash
+# Check a specific README for issues
+node docs-linter/src/cli.js check path/to/README.md
+
+# Get quality score
+node docs-linter/src/cli.js validate path/to/README.md
+
+# Auto-fix safe issues
+node docs-linter/src/cli.js fix path/to/README.md
+```
+
+The docs-linter checks for KM documentation standards including heading capitalization, list formatting, link phrasing, and technical accuracy.
+

.gitignore

@@ -4,6 +4,7 @@
 
 # Claude Code
 .claude/
+!.claude/skills/
 
 # Dependency directories
 node_modules/

cap/README.md

@@ -23,4 +23,4 @@ All CAP projects were generated using the steps outlined in this [blog post](htt
 For more information on integrating CI/CD into your CAP deployment strategy, please refer to the [Continuous Integration and Delivery YouTube Tutorial](https://www.youtube.com/watch?v=gvWSHSZFPok).
 
 ### License
-Copyright (c) 2009-2026 SAP SE or an SAP affiliate company. This project is licensed under the Apache Software License, version 2.0 except as noted otherwise in the [LICENSE](/LICENSES/Apache-2.0.txt) file.
+Copyright (c) 2009-2026 SAP SE or an SAP affiliate company. This project is licensed under the Apache Software License, version 2.0 except as noted otherwise in the [LICENSE](../LICENSES/Apache-2.0.txt) file.