[kenjudy]

Summary

• Live run (node local-code-metrics.js --history squashed) showed runSemanticDuplicateAnalysis returning prose ("Looking at...") instead of JSON, causing parse failure and status: unmeasured
• Root cause: DUPLICATE_SYSTEM_PROMPT was missing the "no other text" clause that CLAUDE_SYSTEM_PROMPT and CFP_SYSTEM_PROMPT both carry
• Adds no other text to the duplicate prompt's JSON instruction
• Adds a prompt content test (DUPLICATE_SYSTEM_PROMPT instructs model to return no other text alongside the JSON)
• Adds scripts/eval-prompts.js: a manual smoke-test that runs the CFP and duplicate prompts against the live API and asserts the responses are parseable JSON matching the expected schema; run before releasing any prompt change (ANTHROPIC_API_KEY=... node scripts/eval-prompts.js)

The commit analysis prompt (CLAUDE_SYSTEM_PROMPT) already carried "no other text" and has not been observed to fail; not touched here.

Test plan

• __tests__/claudeAnalysis.test.js -- 1 new test: DUPLICATE_SYSTEM_PROMPT instructs model to return no other text alongside the JSON
• 824/824 tests pass
• node scripts/eval-prompts.js passes against live API (verified locally: 2 passed, 0 failed)

Generated with Claude Code

[github-actions]

PR Analysis

Size: small (based on production code)
Production Code: 98 lines (2 files)
Test Code: 11 lines (1 files)
Total: 109 lines (3 files)
Test-to-Production Ratio: 0.11:1
No size or quality concerns.

Commit Analysis

Total Commits: 1
Average Commit Size: 98 production lines
Average Files per Commit: 3.0

Metric	Value
Large commits (>100 prod lines)	0/1 (0%)
Sprawling commits (>5 files)	0/1 (0%)
Test/prod co-change rate	1/1 (100%)
Message quality	1/1 (100%)
Median net additions ratio	0.96
Test-only commits	0
Production-only commits	0

Commit Details

6040604 Ken Judy (8/31/2026)
fix: strengthen duplicate prompt JSON enforcement; add prompt eval script
98 prod lines, 3 files [test+prod]

Test Coverage

Test-to-Production Ratio: 0.11:1 (no calibrated target for this ratio; reported descriptively)

DORA Capability Assessment

Archetype: harmonious-high-achiever
Strong batch discipline, test coverage, and message quality. AI tools likely amplifying positive outcomes.

Capability	Metric	Value	Target
Small Batches	Large commit %	0%	<18% (no critical band established)
Small Batches	Sprawling commit %	0%	<18% (no critical band established)
Version Control	Test/prod co-change rate	100%	>23% (no critical band established)
Version Control	Message quality	100%	no band, see spec
AI Risk Signal	Net additions ratio (median)	0.96	no band, see spec
Sample size (1 commit) is below the 5-commit convention for a rate verdict (code-quality-metrics-w9f): large-commit %, sprawling-commit % and test/prod co-change rate are reported below without a pass/fail judgement.
Automated by Code Metrics Workflow

[github-actions]

PR Analysis

Size: large (based on production code)
Production Code: 126 lines (3 files)
Test Code: 11 lines (1 files)
Total: 137 lines (4 files)
Test-to-Production Ratio: 0.09:1
No size or quality concerns.

Commit Analysis

Total Commits: 2
Average Commit Size: 63 production lines
Average Files per Commit: 2.0

Metric	Value
Large commits (>100 prod lines)	0/2 (0%)
Sprawling commits (>5 files)	0/2 (0%)
Test/prod co-change rate	1/2 (50%)
Message quality	2/2 (100%)
Median net additions ratio	0.63
Test-only commits	0
Production-only commits	1

Commit Details

6040604 Ken Judy (8/31/2026)
fix: strengthen duplicate prompt JSON enforcement; add prompt eval script
98 prod lines, 3 files [test+prod]

0ba0347 Ken Judy (8/31/2026)
docs: update README for CFP estimation, eval script, and greenfield n=6
28 prod lines, 1 files

Test Coverage

Test-to-Production Ratio: 0.09:1 (no calibrated target for this ratio; reported descriptively)

DORA Capability Assessment

Archetype: harmonious-high-achiever
Strong batch discipline, test coverage, and message quality. AI tools likely amplifying positive outcomes.

Capability	Metric	Value	Target
Small Batches	Large commit %	0%	<18% (no critical band established)
Small Batches	Sprawling commit %	0%	<18% (no critical band established)
Version Control	Test/prod co-change rate	50%	>23% (no critical band established)
Version Control	Message quality	100%	no band, see spec
AI Risk Signal	Net additions ratio (median)	0.63	no band, see spec
Sample size (2 commits) is below the 5-commit convention for a rate verdict (code-quality-metrics-w9f): large-commit %, sprawling-commit % and test/prod co-change rate are reported below without a pass/fail judgement.
Automated by Code Metrics Workflow