feat: implement dogfood report suggestions

[carlos-alm]

Summary

• Add excludeTests config option (query.excludeTests in .codegraphrc.json) with --include-tests CLI override on all commands
• Add coupling score column (=NNN) to map command output
• Add --depth <n> option to explain for recursive dependency explanation
• Move engine status messages from stdout to stderr via logger
• Fix missing #!/bin/sh shebangs on all husky hooks (Windows compatibility)
• Update DOGFOOD-REPORT-2.1.0 marking all 5 suggestions as implemented

Test plan

• npm run lint — clean
• Integration tests (queries + CLI) — 67/67 passed
• codegraph where explain -T — excludes test files
• codegraph where explain --include-tests — includes test files
• codegraph explain buildGraph --depth 1 -T — recursively explains callees
• codegraph explain openDb --depth 1 -T -j — JSON output with depth works
• codegraph map -T — shows =NNN coupling column
• Husky hooks execute on Windows (pre-commit, commit-msg, pre-push)

[claude]

Claude finished @carlos-alm's task —— View job

---

PR Review: feat: implement dogfood report suggestions

Phase 1: Root Cause Analysis

• Read CLAUDE.md and understand dogfooding context
• Read the dogfood report to understand what problems were identified
• Analyze what specific issues this PR addresses

Phase 2: Configuration Change Detection

• Check for any biome/linting rule changes
• Verify no test timeouts or coverage thresholds were lowered
• Review build configuration changes

Phase 3: Test Integrity

• Examine test modifications for weakened assertions
• Check for removed or conditional test validation
• Verify test coverage for new functionality

Phase 4: Code Quality Review

• Review implementation in cli.js for new options
• Analyze queries.js changes for coupling scores and explain depth
• Check config.js for excludeTests option
• Review builder.js stderr logging changes
• Examine husky hook shebang fixes

Final Assessment

• Provide rating and recommendation

Root Cause Analysis

Problem Being Solved: Five specific usability gaps identified through comprehensive dogfooding of codegraph v2.1.0 on itself. These are legitimate issues: missing CLI options that existed in the data layer, inconsistent output formatting, Windows compatibility problems, and stdout pollution.

Why This Approach: All fixes are targeted and conservative. No major architectural changes, just connecting existing functionality to the CLI and improving output consistency.

Risk Assessment: Very low risk. Changes are primarily additive CLI options, display formatting, and fixing missing shebangs.

Critical Concerns

None identified. This is a clean, well-implemented PR that addresses real usability issues without introducing technical debt or risky configuration changes.

Positive observations:

• Excellent precedence handling in resolveNoTests() helper
• Proper cycle detection in recursive explain implementation
• Consistent application of new CLI options across all affected commands
• Clean separation of concerns (data layer unchanged, only CLI and display improved)
• All status messages properly moved to stderr

Final Recommendation

• Rating: ⭐⭐⭐⭐☆ (4/5)
• Action: APPROVE
• Reasoning: This is a high-quality implementation of legitimate feature requests from real usage. The code is clean, consistent, and follows established patterns. The only reason it's not 5 stars is that it's incremental improvements rather than groundbreaking functionality, but that's exactly what these dogfood suggestions should be.

Recommendation: Approve and merge. This exemplifies good dogfooding practice - identifying real pain points through usage and addressing them systematically.

---

[greptile-apps]

Greptile Summary

This PR implements all 5 suggestions from DOGFOOD-REPORT-2.1.0: adds excludeTests config option with --include-tests CLI override, adds coupling score column to map output, adds --depth option to explain for recursive dependency exploration, migrates engine status messages from stdout to stderr, and fixes missing shebangs on husky hooks for Windows compatibility.

Key changes:

• Config system now supports query.excludeTests default with CLI override via --include-tests flag (applied to all 17 commands)
• map command displays coupling scores as =NNN column showing combined fan-in/fan-out
• explain command supports --depth <n> to recursively explain callees up to N levels with cycle-safe tracking
• All status messages in builder.js migrated from console.log to logger.info() (stderr)
• Fixed Windows compatibility by adding #!/bin/sh shebangs to all three husky hooks
• Renamed --include-tests to --with-test-source on context command to avoid naming collision

All integration tests pass (67/67), test spies correctly updated to check stderr instead of stdout.

Confidence Score: 5/5

• This PR is safe to merge with minimal risk
• All changes are well-tested with 67/67 integration tests passing, the implementation follows existing patterns, backward compatibility is maintained (config defaults to false, preserving existing behavior), and the test spies were correctly updated to match the stdout-to-stderr migration
• No files require special attention

Important Files Changed

Filename	Overview
src/builder.js	migrated status messages from stdout to stderr using logger.info
src/cli.js	added excludeTests config support, --include-tests flag, resolveNoTests helper, --depth for explain, renamed --include-test-source to --with-test-source
src/config.js	added query.excludeTests: false to DEFAULTS config
src/queries.js	added coupling field to moduleMapData, =NNN column to map output, --depth support for explain with recursive callee expansion
tests/integration/build.test.js	updated test spies from console.log to stderr.write after status message migration

_{Last reviewed commit: 242066f}

[greptile-apps]

_{8 files reviewed, 1 comment}

_{Edit Code Review Agent Settings | Greptile}

[greptile-apps]

Consider renaming --include-test-source to something like --with-test-source to distinguish it from --include-tests (which controls test file filtering, not test source inclusion)

_{Note: If this suggestion doesn't match your team's coding style, reply to this and let me know. I'll remember it for next time!}

[claude]

Claude finished @carlos-alm's task —— View job

---

PR Review: feat: implement dogfood report suggestions

Phase 1: Root Cause Analysis

• Read CLAUDE.md and understand dogfooding context
• Read the dogfood report to understand what problems were identified
• Analyze what specific issues this PR addresses

Phase 2: Configuration Change Detection

• Check for any biome/linting rule changes
• Verify no test timeouts or coverage thresholds were lowered
• Review build configuration changes

Phase 3: Test Integrity

• Examine test modifications for weakened assertions
• Check for removed or conditional test validation
• Verify test coverage for new functionality

Phase 4: Code Quality Review

• Review implementation in cli.js for new options
• Analyze queries.js changes for coupling scores and explain depth
• Check config.js for excludeTests option
• Review builder.js stderr logging changes
• Examine husky hook shebang fixes

Final Assessment

• Provide rating and recommendation

Root Cause Analysis

Problem Being Solved: Five specific usability gaps identified through comprehensive dogfooding of codegraph v2.1.0 on itself. These are legitimate issues: missing CLI options that existed in the data layer, inconsistent output formatting, Windows compatibility problems, and stdout pollution.

Why This Approach: All fixes are targeted and conservative. No major architectural changes, just connecting existing functionality to the CLI and improving output consistency.

Risk Assessment: Very low risk. Changes are primarily additive CLI options, display formatting, and fixing missing shebangs.

Critical Concerns

None identified. This is a clean, well-implemented PR that addresses real usability issues without introducing technical debt or risky configuration changes.

Positive observations:

• Excellent precedence handling in resolveNoTests() helper
• Proper cycle detection in recursive explain implementation
• Consistent application of new CLI options across all affected commands
• Clean separation of concerns (data layer unchanged, only CLI and display improved)
• All status messages properly moved to stderr

Final Recommendation

• Rating: ⭐⭐⭐⭐☆ (4/5)
• Action: APPROVE
• Reasoning: This is a high-quality implementation of legitimate feature requests from real usage. The code is clean, consistent, and follows established patterns. The only reason it's not 5 stars is that it's incremental improvements rather than groundbreaking functionality, but that's exactly what these dogfood suggestions should be.

Recommendation: Approve and merge. This exemplifies good dogfooding practice - identifying real pain points through usage and addressing them systematically.

---

[carlos-alm]

@greptileai

[claude]

Claude finished @carlos-alm's task —— View job

---

PR Review: feat: implement dogfood report suggestions

Phase 1: Root Cause Analysis

• Read CLAUDE.md and understand dogfooding context
• Read the dogfood report to understand what problems were identified
• Analyze what specific issues this PR addresses

Phase 2: Configuration Change Detection

• Check for any biome/linting rule changes
• Verify no test timeouts or coverage thresholds were lowered
• Review build configuration changes

Phase 3: Test Integrity

• Examine test modifications for weakened assertions
• Check for removed or conditional test validation
• Verify test coverage for new functionality

Phase 4: Code Quality Review

• Review implementation in cli.js for new options
• Analyze queries.js changes for coupling scores and explain depth
• Check config.js for excludeTests option
• Review builder.js stderr logging changes
• Examine husky hook shebang fixes

Final Assessment

• Provide rating and recommendation

Root Cause Analysis

Problem Being Solved: Five specific usability gaps identified through comprehensive dogfooding of codegraph v2.1.0 on itself. These are legitimate issues: missing CLI options that existed in the data layer, inconsistent output formatting, Windows compatibility problems, and stdout pollution.

Why This Approach: All fixes are targeted and conservative. No major architectural changes, just connecting existing functionality to the CLI and improving output consistency.

Risk Assessment: Very low risk. Changes are primarily additive CLI options, display formatting, and fixing missing shebangs.

Critical Concerns

None identified. This is a clean, well-implemented PR that addresses real usability issues without introducing technical debt or risky configuration changes.

Positive observations:

• Excellent precedence handling in resolveNoTests() helper (src/cli.js:60-64)
• Proper cycle detection in recursive explain implementation (src/queries.js:1954-1985)
• Consistent application of new CLI options across all affected commands
• Clean separation of concerns (data layer unchanged, only CLI and display improved)
• All status messages properly moved to stderr (verified in src/builder.js)

Final Recommendation

• Rating: ⭐⭐⭐⭐☆ (4/5)
• Action: APPROVE
• Reasoning: This is a high-quality implementation of legitimate feature requests from real usage. The code is clean, consistent, and follows established patterns. The only reason it's not 5 stars is that it's incremental improvements rather than groundbreaking functionality, but that's exactly what these dogfood suggestions should be.

Recommendation: Approve and merge. This exemplifies good dogfooding practice - identifying real pain points through usage and addressing them systematically.

---

[greptile-apps]

_{9 files reviewed, no comments}

_{Edit Code Review Agent Settings | Greptile}