Prompt file imported from pawansinghal007-bot/strategic-learning-unified-theatre (.github/prompts/plan-cleanupAndCoveragePlan.prompt.md). Copyright stays with the author.

Plan: Project Cleanup & Structure Audit

TL;DR: The project root is severely cluttered with 100+ backup directories, timestamped snapshots, temporary debug files, and output artifacts. These should be moved to a .archive/ directory or deleted. The core project structure is intact but obscured by noise. Cleanup will: improve navigation, reduce disk space (~500MB+), and maintain referential integrity via git history.

Critical Findings

🔴 High-Priority Clutter (Delete or Archive)

Snapshot Backups (100+ directories)

• strategic-learning-unified-theatre-ai-snapshot-sprint*-* (50+ dirs)
• strategic-learning-unified-theatre-ai-snapshot-v*.*. (15+ dirs)
• strategic-learning-unified-theatre-ai-snapshot-v2.*-stable (10+ dirs)
• These appear to be periodic AI/LLM context snapshots; they're no longer needed in active development

Architecture Baseline Files (30+ timestamped backups)

• PROJECT_ARCHITECTURE_BASELINE-*.md (30 files with timestamps)
• PROJECT_ARCHITECTURE_BASELINE.md (keep one canonical version)
• These are historical snapshots; only keep the latest or one reference version

Debug & Diagnostic Files (temporary)

• diagnose.cjs, diagnose2.cjs + Zone.Identifier variants
• fix-dashboard*.{cjs,js} (5+ variants)
• fix-ipc.cjs
• debug-plugin-run.mjs
• test-bc2-integration.js, test-bc2-diagnostics.js, test-llama.mjs, test_gpu.py

Sprint & Report Files (obsolete data)

• ai-snapshot-sprint*.md (3 files)
• sprint*.csv, sprint*.txt (Sonar-related)
• impacted-tests.txt, remaining_*.txt
• Sonar audit reports: sonar_issues.csv, sonar_hotspots.csv, sonar_reliability_maintainability.csv, sonar_issues.md

Audit & Log Files

• audit-log-ws-*.json, audit-log-ws-*.html, audit-log-ws-*.txt
• project_audit_dump.txt, git-history.txt
• key-files.txt, source-files.txt, src-structure.txt, tsconfigs.txt, root-package.txt, vite-config.txt

Other Temporary Files

• hello.py, test_gpu.py, test-error.ts (test stubs)
• test_summary.txt, tmp-test-*, smoke-test-sprint12.js
• BC2-INTEGRATION-GUIDE.js
• *.tar.gz, sprint52-input-files.tar/
• .continuerc.json (Continue IDE config, can be ignored)
• run-tests.cjs, smoke-test-sprint12.js, test-*.js variants
• load-env.ps1, safe_audit.sh, search_repo.py
• Modelfile (Ollama model file; only needed if AI runtime is part of active development)

Build Output & Generated Files

• build/ (rebuild as needed)
• dist/ (rebuild as needed)
• coverage/ (regenerate per test run)
• playwright-report*/ (regenerate per test run)
• test-results/ (regenerate per test run)
• robot-results/ (regenerate per test run)
• dumps/, audit_chunks/ (temporary processing)

Generated Reports

• sonar-project.properties, sonar-issues-report.json, sonar_*.csv, sonar_*.json
• report-task.txt, sprint*-report.md

🟡 Medium-Priority Issues (Verify & Standardize)

Multiple Config Variants for Same Tool

• Playwright: playwright.config.ts + playwright.human.config.{cjs,js} + playwright.ui.config.{cjs,js}
  • Decision: Keep only needed variants; verify which are actually used in CI/CD and tests
• Vitest: vitest.config.{js,ts} + vitest.integration.config.js + .log file
  • Decision: Keep all if used; remove .log

Duplicate / Variant Files

• browser-bridge.js + browser-bridge.jsy (likely a typo)
• diagnose.cjs:Zone (2).Identifier (Windows zone marker; should be .gitignored)

Untracked Directories (not in .gitignore but should be)

• .codex/, .continue/ (IDE integration caches)
• .scannerwork/ (SonarQube working directory)
• reports/, robot/ (should review if these are build artifacts or source)

🟢 Legitimate Project Files (Keep)

Core Source

• src/ — Main application source
• renderer/ — Renderer process code
• plugins/ — Plugin system
• electron-tray/, electron-ui/ — Electron UI code
• Service/, Preload/, IPC/, main/ — Entry points and IPC modules

Testing & Validation

• tests/, test/, e2e/, e2e-design/, robot/ — Test suites (verify robot/ is not just generated output)
• playwright.*.config.* — Test configuration (if all are used)
• vitest.*.config.* — Unit test configuration

Documentation & Configuration

• .github/ — GitHub workflows, instructions
• .vscode/ — Workspace settings
• scripts/ — Build and utility scripts
• docs/ — Documentation
• config/ — Configuration files
• schema.json — Data schema
• .env — Environment variables
• package.json, package-lock.json — Dependencies
• tsconfig.json — TypeScript config
• vite.config.js, postcss.config.cjs — Build configs
• tailwind.config.js — Styling
• .git/, .gitignore — Version control

Electron Build Artifacts (can rebuild but slow)

• release/ — Packaged releases (keep if not easily regenerated; consider archiving)

Infrastructure

• node_modules/ — Dependencies (rebuild from lock file)
• .scannerwork/ — SonarQube cache (can regenerate)
• vscode-extension/ — VS Code extension source (if active project)

Cleanup Plan

Phase 1: Automated Archive & Delete (Safe)

1. Create .archive/ directory in root
2. Move all strategic-learning-unified-theatre-ai-snapshot-* directories → .archive/ai-snapshots/
3. Move all PROJECT_ARCHITECTURE_BASELINE-*.md (keep latest 1-2) → .archive/architecture-baselines/
4. Delete all temporary debug files (diagnose*, fix-dashboard*, etc.)
5. Delete all sprint audit CSVs and reports (sprint*.csv, sonar_*.csv, sonar-issues*, etc.)
6. Delete all temporary logs and text files (audit-log-ws-*, git-history.txt, key-files.txt, etc.)
7. Delete temporary test files (hello.py, test-error.ts, test_summary.txt, etc.)

Phase 2: Build Artifact Cleanup

1. Delete build/, dist/, coverage/, playwright-report*/, test-results/, robot-results/
   • These regenerate automatically on test/build runs
2. Delete dumps/, audit_chunks/ (temporary processing)
3. Delete generated reports (report-task.txt, sonar-issues-report.json, etc.)

Phase 3: Configuration & Variant Review

1. Playwright configs: Test each variant in CI to identify which are actually used
   • If only 1-2 are used, delete the unused ones
   • If all used, keep; document why each exists
2. Vitest configs: Verify vitest.integration.config.js is used; delete if not
3. Debug logs: Remove vitest.log

Phase 4: .gitignore Alignment (Prevent Recurrence)

1. Add to .gitignore:
   • .codex/, .continue/
   • build/, dist/, coverage/
   • *.db-shm, *.db-wal (already there, verify)
   • *-output.txt, *-output.json (already there, verify)
   • Sonar temp files patterns
2. Remove from tracked git history (if small enough):
   • Historic snapshots via git rm -r --cached for .archive/
   • Verify this doesn't break referential integrity

Phase 5: Verify Integrity

1. Run test suite: npm test
2. Run integration tests: npm run test:integration
3. Run Playwright tests (at least one variant): npm run test:ui
4. Verify Electron build: npm run electron:dev
5. Check CI/CD workflows still work (review .github/workflows/)

Disk Space Impact

Before: ~500MB+ (estimated)

• 100+ snapshot dirs: ~300-400MB
• 30+ architecture baselines: ~5MB
• Debug & temp files: ~50MB
• Build artifacts & reports: ~50-100MB

After: ~50-100MB reduction

• Keep only .archive/ as historical reference (compress if needed)

Files to Review (Interactive Decisions)

Before deletion, confirm with user:

1. Modelfile — Is Ollama/LLM runtime part of active development?
2. robot/ & robot.config.json — Are Robot Framework tests actively maintained or can they be archived?
3. BC2-INTEGRATION-GUIDE.js — Is BC2 integration still active?
4. vscode-extension/ — Is this a standalone project or archived?
5. release/ — Are old releases needed or can they be archived?

ADDITIONAL FINDING: Sonar Coverage < 100%

Root Causes Identified

1. CRITICAL ISSUE: Incomplete Coverage Include List

The vitest.config.ts (and vitest.config.js) only includes 6 specific files in coverage measurement:

include: [
  'src/accounts/secret-store.js',
  'src/daemon/daemon-runner.js',
  'src/browser-bridge.js',
  'src/agent-handoff.js',
  'src/llm/local-llm.js',
  'src/idea-store.js',
]

Actual src/ contains:

• 40+ top-level module files (.js, .ts)
• 20+ subdirectories with additional code
• At least 70+ source files NOT covered

Impact: ~95% of codebase is excluded from coverage measurement → Reports show <100% because most code isn't being measured at all.

2. Coverage Threshold Mismatch

• Configured threshold: 70% (statements, branches, functions, lines)
• Goal: 100% coverage
• Reality: Only 6 files are measured, so threshold of 70% on 6 files ≠ 100% project-wide

3. Test File Distribution Problem

• Test discovery looks in: tests/, src/, electron-ui/, renderer/, e2e/
• But coverage only includes 6 src files
• No evidence of comprehensive test files for each module

Why Coverage Was Deliberately Limited

The configuration suggests intentional coverage scoping to critical files:

• Secret store (security-sensitive)
• Daemon runner (core service)
• Browser bridge (main integration point)
• Agent handoff (state management)
• Local LLM (AI integration)
• Idea store (data persistence)

This is a valid pattern IF:

• These 6 files are the only ones requiring 100% coverage
• Other files are covered by integration/e2e tests
• The project acknowledges partial coverage as acceptable

But SonarQube reports show coverage < 100%, which suggests one of: A. Full coverage was the goal but implementation was incomplete B. SonarQube is configured to measure all src/ but vitest only covers 6 files C. Missing test files for covered modules

How SonarQube Reads Coverage

1. Takes LCOV report from: coverage/lcov.info (vitest output)
2. LCOV file only contains lines for 6 covered files (because vitest.config limits include)
3. SonarQube's Sonar.sources = src (all src/ files)
4. SonarQube compares: "Coverage report has 6 files, but src/ has 70+ files"
5. Result: Coverage % = (lines covered in 6 files) / (total lines in src/) = < 100%

Solution Options

Option A: Expand Coverage to All Source Code

• Add all src files to coverage.include in vitest.config
• Write/fix tests for all modules
• Expect time investment: HIGH (~40-80 hours for comprehensive tests)
• Pros: True 100% coverage visibility
• Cons: May expose test infrastructure gaps

Option B: Exclude Non-Critical Files from SonarQube

• Keep 6-file coverage in vitest
• Add sonar.coverage.exclusions patterns to exclude 70+ src files from SonarQube measurement
• Pros: Reasonable effort, honest reporting
• Cons: SonarQube may still flag as incomplete

Option C: Split Coverage Strategy (Recommended)

• Tier 1 (100% coverage required): 6 critical files (already configured)
• Tier 2 (70%+ coverage): Add 15-20 important business logic files
• Tier 3 (excluded): Infrastructure, utilities, config, generated files
• Update SonarQube to exclude Tier 3
• Expected time: 20-30 hours
• Pros: Balanced coverage with achievable goals

Option D: Keep Current Setup, Update Documentation

• Document that coverage is intentionally limited to 6 critical files
• Update SonarQube quality gate to expect ~15-25% project coverage (honest baseline)
• Keep investigation for future work
• Time: 1-2 hours
• Pros: Quick, acknowledges current state
• Cons: May fail CI/CD if quality gate expects higher %

Execution Roadmap

Decision Required

Archive Strategy:

• Option A: Move to .archive/, commit, push (keeps history visible)
• Option B: Delete locally, keep in git history (cleaner root, same recovery potential)
• Option C: Move to separate historical branch (advanced; separate history timeline)

Recommendation: Option A — .archive/ directory keeps items findable and is low-friction.

Coverage Decision Matrix

Final Recommendation for Coverage: Option C (Tiered) offers the best path forward—expand coverage incrementally without overwhelming effort, while maintaining quality gates on critical files.