Daily Code Metrics Report - 2026-03-10

[github-actions[bot]]

The amplihack codebase was analyzed on 2026-03-10. The repository contains 1,543,094 total lines across all tracked file types, with Python (669,489 LOC) and Markdown (820,813 LOC) dominating the distribution. The overall quality score is 72.3/100 — a notable change from yesterday's 81.1, primarily driven by a significant increase in code churn (74 commits, 902 files modified, +53,720 net LOC) which lowers churn stability. Test coverage remains healthy at 0.769 test-to-source ratio.

The most significant finding is the explosive growth: LOC increased by ~131K lines day-over-day, with major additions in src/amplihack/agents/goal_seeking/, experiments/hive_mind/, and fleet tests. This active development is healthy but is temporarily depressing the churn stability score. The comment density (6.5%) and code organization score (334 large files > 500 LOC) represent areas for sustained improvement.

📊 Visualizations

LOC Distribution by Language

Markdown dominates at 53.2% of total LOC, reflecting amplihack's heavy documentation and workflow-as-code model. Python is the primary source language at 43.4%. Rust, TypeScript, and JavaScript contribute a small but growing footprint.

Top Directories by LOC

.claude (403,945) and docs (390,422) are the two largest directories, each exceeding amplifier-bundle (260,536) and src (190,692). The tests directory (182,509 LOC) reflects strong test investment. The src + tests balance is healthy.

Quality Score Breakdown

The 72.3/100 quality score reflects strong documentation (16/20) and test coverage (25.4/30), but is constrained by high churn this week (8.9/15) and low comment density (6.0/10). Code organization (16.0/25) is impacted by 334 files exceeding 500 LOC.

Test Coverage Analysis

Test LOC (291,045) vs source LOC (378,444) yields a 0.769 ratio — within the target range of 0.5–1.0. This is excellent for an actively developed AI framework. 764 test files covering 2,140 source files indicates broad coverage intent.

Code Churn (Last 7 Days)

74 commits touched 902 source files with +142,369 / -88,649 lines (net +53,720). The most churned files are power_steering_checker.py (refactored), learning_agent.py (new addition), and _hierarchical_memory_local.py (new). Lock file churn (33 files, +9,458/-7,295 lines) is tracked separately and excluded from quality scoring.

Historical Trends (2 Days)

Only 2 data points so far (Mar 9–10). Day-over-day LOC grew +131K (+9.3%). Quality score dropped from 81.1 → 72.3 due to churn spike. Trends will become more meaningful with 7+ days of history.

📈 Detailed Metrics

Size Metrics

Lines of Code by Language

Language	LOC	% of Total	Change (1d)
Markdown	820,813	53.2%	➡️ +0.4%
Python	669,489	43.4%	⬆️ +30.5%
YAML	39,821	2.6%	⬇️ -36.6%
JSON	7,337	0.5%	⬆️ +136.8%
JavaScript	3,413	0.2%	n/a
Rust	1,600	0.1%	new
TypeScript	264	<0.1%	⬇️ -92.8%
TOML	357	<0.1%	➡️ 0%

Lines of Code by Directory

Directory	LOC	% of Total	Files
.claude	403,945	26.2%	~1,200+
docs	390,422	25.3%	688
amplifier-bundle	260,536	16.9%	~500+
src	190,692	12.4%	~700+
tests	182,509	11.8%	~300+
Specs	44,736	2.9%	~100+
scripts	14,209	0.9%	~50+
experiments	12,225	0.8%	~30+

Quality Indicators

• Average Python File Size: 312 lines
• Large Files (>500 LOC): 334 files
• Python Function Count: 25,429 functions
• Comment Lines: 44,090 lines (6.5% ratio)
• Comment Density: 6.5%

Test Coverage

• Test Files: 764 files
• Test LOC (test_lines_of_code): 291,045 lines
• Source LOC: 378,444 lines
• Test-to-Source Ratio (test_to_source_ratio): 0.769
• Trend (1d): ⬇️ from 1.062 (ratio normalized as source LOC grew faster)

Code Churn (Last 7 Days)

Source Code Churn (Excludes *.lock.yml)

• Files Modified: 902 files
• Commits: 74 commits
• Lines Added: +142,369 lines
• Lines Deleted: -88,649 lines
• Net Change: +53,720 lines

Most Active Source Files

1. .claude/tools/.../power_steering_checker.py: -5,200 lines (refactored away)
2. amplifier-bundle/.../power_steering_checker.py: -5,035 lines (refactored away)
3. src/.../learning_agent.py: +3,103 lines (new feature)
4. amplifier-bundle/.../main_checker.py: +1,278/-1,278 lines (churn)
5. src/.../_hierarchical_memory_local.py: +2,362 lines (new feature)
6. docs/.../FINAL_SYNTHESIS_MEMORY.md: -1,950 lines (doc cleanup)
7. docs/.../KNOWLEDGE_GRAPH_RESEARCH.md: -1,901 lines (doc cleanup)
8. src/.../test_fleet_cli.py: +1,783 lines (new tests)
9. experiments/.../run_20agent_kuzu_eval.py: +1,713 lines (experiment)
10. experiments/.../run_full_distributed.py: +1,587 lines (experiment)

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

• Lock Files Modified: 33 files
• Lines Added: +9,458 lines
• Lines Deleted: -7,295 lines
• Net Change: +2,163 lines

Note: Lock file churn is tracked separately and excluded from quality score calculations.

Workflow Metrics

• Total Workflow Files (.md) (total_workflows): 56 files in .github
• Compiled Workflows (.lock.yml): 17 files
• GitHub Actions .yml: 35 workflows
• *.claude .md files: 590 files (skills, agents, commands, context)
• Growth (1d): ⬆️ +17 new GitHub Actions workflows

Documentation

• Doc Files (docs/): 688 files
• Doc LOC: 294,944 lines
• Code-to-Docs Ratio: 2.26:1
• Python LOC: 669,489 → docs provide 44% as much content as Python code

Quality Score: 72.3/100

Component Breakdown

Component	Score	Max	Ratio
Test Coverage (30%)	25.4	30	84.7%
Code Organization (25%)	16.0	25	64.0%
Documentation (20%)	16.0	20	80.0%
Churn Stability (15%)	8.9	15	59.3%
Comment Density (10%)	6.0	10	60.0%
Total	72.3	100	72.3%

Day-over-Day Comparison

Metric	2026-03-09	2026-03-10	Change
Total LOC	1,412,011	1,543,094	⬆️ +9.3%
Python LOC	513,240	669,489	⬆️ +30.5%
Test Ratio	1.062	0.769	⬇️ -27.6%
Quality Score	81.1	72.3	⬇️ -8.8pts
Commits (7d)	62	74	⬆️ +19.4%
Files Changed	862	902	⬆️ +4.6%
Functions	19,459	25,429	⬆️ +30.7%

💡 Insights & Recommendations

1. Reduce large file count (334 files > 500 LOC): The code organization score (64%) is the second-weakest component. Prioritize splitting learning_agent.py (3,103 LOC), _hierarchical_memory_local.py (2,362 LOC), cli.py (1,949 LOC), and auto_mode.py (1,873 LOC) into focused modules. Target: reduce to <250 large files.

2. Improve comment density (6.5% → target 12-15%): The comment density score (60%) is the joint-weakest component. New code being added (learning_agent.py, _hierarchical_memory_local.py) should include inline documentation. Focus first on public APIs and complex algorithms.

3. Monitor churn stabilization: The current 74-commit/7-day velocity (+53K net lines) reflects active feature development (goal-seeking agents, hierarchical memory). Churn stability should recover as this sprint concludes. Watch for the score to rebound toward 12+/15 over the next 7 days.

4. Test ratio recovery: The test-to-source ratio dropped from 1.062 to 0.769 as source code grew faster than tests. New files learning_agent.py, _hierarchical_memory_local.py need test coverage. Aim to add tests within 48 hours of new source additions.

5. Archive or remove experimental files: experiments/hive_mind/ contains 12,225 LOC of one-off evaluation scripts that inflate churn and file counts. Consider archiving completed experiments to a separate branch or tagging them as archived.

---

Report generated by Daily Code Metrics workflow
Historical data: 2 days | Last updated: 2026-03-10 UTC
Workflow run: 22913534363

AI generated by Daily Code Metrics and Trend Tracking Agent · history

• expires on Mar 13, 2026, 4:53 PM UTC

[github-actions[bot]]

This discussion was automatically closed because it expired on 2026-03-13T16:53:56.579Z.

Closed by Workflow