Daily Code Metrics Report - 2026-02-20

[github-actions[bot]]

The gh-aw codebase has grown to 821,420 total LOC across 7 languages, with Go remaining the dominant language at 430,218 lines (52.4%). The quality score is 78/100, unchanged from recent runs, with excellent test coverage (2.33x ratio) and strong documentation offset by high code churn volume this week.

The 7-day churn window shows 1,117 source files modified across 455 commits, producing a net gain of +58,348 lines. The most active areas include the docs editor, schema files, and skill documentation. Churn stability scores at 3/15 — the highest volume of change in recent weeks — reflecting active feature development rather than decay. Lock files (157 files, ~130K lines changed) continue to be tracked separately to isolate this noise from quality calculations.

The workflow growth is also notable: 210 workflow markdown files, with the .github/workflows/ directory now spanning 256,456 lines across 388 files.

📊 Visualizations

LOC Distribution by Language

Go dominates at 52.4% of total LOC (430K lines), followed by YAML at 24.3% (199K), Markdown at 21.5% (177K), and Shell at 1.1% (9K). TypeScript and JavaScript together represent under 1% of the codebase. The language mix reflects a primarily server-side Go application with significant workflow infrastructure.

Top Directories by LOC

docs/ leads all directories at 873,515 LOC (note: includes generated documentation and public assets). pkg/ is the largest source of application code at 427,320 LOC. The .github/workflows/ directory continues growing with 256,456 lines across compiled and markdown workflow files.

Quality Score Breakdown

Overall quality score: 78/100. Test coverage achieves a perfect 30/30. Code organization scores 20/25 (80%) reflecting 216 large files out of 1,511. Documentation earns 17/20. Churn stability is the primary drag at 3/15 due to very high weekly activity volume. Comment density is healthy at 8/10.

Test Coverage Analysis

Test LOC (300,929) exceeds source LOC (129,289) by 2.33x, placing the codebase firmly in the "Excellent" range. Test files number 958 compared to ~553 non-test Go files. This ratio has been stable across multiple measurement cycles.

Code Churn - Last 7 Days

The highest-churn source files this week: actions/setup/js/copilot-client.js (+5,491/-5,491), docs/public/editor/index.html (+2,193/-1,901), and pkg/parser/schemas/main_workflow_schema.json (+2,056/-1,845). The safe-outputs specification doc grew by +3,806 lines net with minimal deletions, suggesting significant new documentation. All *.lock.yml generated files are excluded from this chart.

Historical Trends - 30 Days

The total LOC has been on a steady upward trajectory from ~800K in early January to ~821K today. The test-to-source ratio has remained stable above 2.0x. Quality scores have generally held at 78-82/100 over the tracking period. Note: 5 data points with quality score data are available for the trend chart; the full 23-entry history spans back to January 7.

📈 Detailed Metrics

Size Metrics

Lines of Code by Language

Language	LOC	Files	% of Total
Go	430,218	1,511	52.4%
YAML	199,518	183	24.3%
Markdown	176,987	751	21.5%
Shell	8,952	61	1.1%
JavaScript	3,195	13	0.4%
TypeScript	1,721	17	0.2%
Python	829	3	0.1%
Total	821,420	2,539	100%

Lines of Code by Directory

Directory	LOC	Files	% of Total
docs/	873,515	207	—
.github/workflows/	256,456	388	—
pkg/	427,320	1,694	—
scratchpad/	37,929	62	—
cmd/	2,232	12	—

Quality Indicators

• Average File Size: 569 lines (Go files)
• Large Files (>500 LOC): 216 files (14.3% of Go files)
• Function Count: 7,589 functions
• Comment Lines: 44,899 lines
• Comment Ratio: 10.4%

Test Coverage

• Test Files: 958 files
• Test LOC (test_lines_of_code): 300,929 lines
• Source LOC: 129,289 lines
• Test-to-Source Ratio (test_to_source_ratio): 2.327
• Status: Excellent (target: ≥1.0)
• vs. 2026-02-19: +0.0 (stable)

Code Churn (Last 7 Days)

Source Code Churn (Excludes *.lock.yml)

• Files Modified: 1,117 files
• Commits: 455 commits
• Lines Added: +100,437 lines
• Lines Deleted: -42,089 lines
• Net Change: +58,348 lines

Most Active Source Files

1. actions/setup/js/copilot-client.js: +5,491/-5,491 (10,982 total)
2. docs/public/editor/index.html: +2,193/-1,901 (4,094 total)
3. pkg/parser/schemas/main_workflow_schema.json: +2,056/-1,845 (3,901 total)
4. docs/src/content/docs/reference/safe-outputs-specification.md: +3,806/-42 (3,848 total)
5. skills/dictation/SKILL.md: +1,063/-1,051 (2,114 total)
6. docs/public/editor/editor.js: +1,265/-797 (2,062 total)
7. docs/public/editor/autocomplete-data.json: +1,760/-222 (1,982 total)
8. specs/artifacts.md: +1,680/-254 (1,934 total)

Workflow Lock File Churn (*.lock.yml only)

• Lock Files Modified: 157 files
• Lines Added: +71,570 lines
• Lines Deleted: -58,985 lines
• Net Change: +12,585 lines

Note: Lock file churn is reported separately and excluded from quality score calculations to avoid noise from generated files.

Workflow Metrics

• Total Workflow Files (.md) (total_workflows): 210 files
• Compiled Workflows (.lock.yml): 153 files
• Total Workflow YML files: 171 files
• Total Files in .github/workflows: 388 files
• Average Workflow MD Size: ~267 lines

Documentation

• Doc Files (docs/): 207 files
• Doc LOC: 873,515 lines
• Code-to-Docs Ratio: 0.49:1 (more docs than code)

Quality Score: 78/100

Component Breakdown

• Test Coverage (30%): 30/30 ✅ — ratio of 2.327x is exceptional
• Code Organization (25%): 20/25 — 14.3% large files, avg 569 LOC/file
• Documentation (20%): 17/20 — strong docs-to-code ratio
• Churn Stability (15%): 3/15 ⚠️ — very high weekly churn (1,117 files)
• Comment Density (10%): 8/10 — 10.4% comment ratio

💡 Insights & Recommendations

1. High churn in docs/editor assets: actions/setup/js/copilot-client.js and multiple docs/public/editor/ files show large bidirectional churn, suggesting these may be build artifacts or auto-generated files. Consider adding them to .gitignore or a separate tracking category to reduce noise.

2. Large file count remains elevated: 216 Go files exceed 500 LOC (14.3%), contributing to the 20/25 code organization score. Targeting files that have grown organically (e.g., pkg parser schemas) for decomposition would improve this score.

3. Safe-outputs specification growing fast: The safe-outputs-specification.md grew by +3,806 lines this week with minimal deletions — this represents significant new API surface documentation worth reviewing for completeness and accuracy.

4. Test coverage is a genuine strength: The 2.33x test-to-source ratio is well above healthy thresholds and has been stable. This is a meaningful signal of codebase maintainability.

5. Workflow infrastructure continues scaling: With 153 lock files and 210 markdown workflows, the workflow layer is substantial. Periodic audits of stale or rarely-triggered workflows could reduce maintenance overhead.

---

Report generated by Daily Code Metrics workflow
Historical data: 23 entries from 2026-01-07 | Last updated: 2026-02-20 UTC
Workflow run: §22222614219

AI generated by Daily Code Metrics and Trend Tracking Agent

• expires on Feb 23, 2026, 11:49 AM UTC