feat: add update-changelog skill

[indiamos]

Adds an update-changelog skill for drafting and writing changelog entries from commits ahead of main.

Detailed changes

• Added .claude/skills/update-changelog/SKILL.md
• Skill reads commits ahead of main, groups user-visible entries under ### Category subheadings matching this project's taxonomy (### New Skills, ### PDCA Framework Redesigns, etc.), and presents a review step before writing
• Entries are written under ## Unreleased (inserting with a --- separator if the section doesn't exist yet, appending if it does); passing a version argument (e.g. v1.2.0 or 1.2.0) writes a versioned heading instead
• CONTRIBUTING.md updated to mention /update-changelog in the PR checklist

[github-actions]

PR Analysis

Size: medium (based on production code)
Production Code: 123 lines (3 files)
Test Code: 0 lines (0 files)
Total: 123 lines (3 files)
Test-to-Production Ratio: 0.00:1

Concerns

• Median net additions ratio 1.00 exceeds 0.50 - possible batch-acceptance pattern

Strengths

• Incremental development pattern detected
• Message quality 100% meets discipline threshold

Commit Analysis

Total Commits: 2
Average Commit Size: 62 production lines
Average Files per Commit: 1.5

Metric	Value
Large commits (>100 prod lines)	1/2 (50%)
Sprawling commits (>5 files)	0/2 (0%)
Test-first discipline	0/2 (0%)
Message quality	2/2 (100%)
Median net additions ratio	1.00
Test-only commits	0
Production-only commits	2

Commit Details

047e119 India Amos (5/27/2026)
feat: add update-changelog skill
120 prod lines, 2 files [large]

4ed6135 India Amos (6/1/2026)
docs: update CHANGELOG.md
3 prod lines, 1 files

Test Coverage

Test Adequacy: poor

• No test changes for production code modifications
• All commits lack test coverage

Target ratio: 0.5-2.0 test lines per production line

DORA Capability Assessment

Archetype: foundational-challenges
Weak testing or batch discipline detected. Consider strengthening practices before scaling AI usage.

Capability	Metric	Value	Target
Small Batches	Large commit %	50%	<20%
Small Batches	Sprawling commit %	0%	<10%
Version Control	Test-first discipline	0%	>50%
Version Control	Message quality	100%	>60%
AI Risk Signal	Net additions ratio (median)	1.00	<0.50
Automated by Code Metrics Workflow