perf: reduce WASM boundary crossings + streamline README

[carlos-alm]

Summary

• JS extractor: reduce WASM boundary crossings by batching tree-sitter node property reads, cutting per-file overhead
• README: restore concise project highlight blockquote, remove per-tool tradeoff table and Narsil-MCP deep-dive while keeping the feature comparison grid and differentiators table

Test plan

• npm test passes (JS extractor changes are behavior-preserving)
• README renders correctly on GitHub (blockquote, feature table, differentiators)

[greptile-apps]

Greptile Summary

Streamlined README to focus on value proposition instead of competitor analysis. Replaced verbose 3-line blockquote with concise "Free. Open source. Fully local" message, and removed 54 lines of detailed competitor tradeoff tables and Narsil-MCP comparison.

Note: PR description mentions "JS extractor WASM boundary crossing" changes, but those were in commit d4ef6da (PR #63) which was already merged. This PR only contains README documentation changes.

• Simplified blockquote highlight from technical explanation to punchy value statement
• Removed "How other tools compare" tradeoff table (10+ tools)
• Removed "Codegraph vs. Narsil-MCP: How to Decide" deep-dive section
• Retained feature comparison grid and differentiators table as intended

Confidence Score: 5/5

• This PR is safe to merge with zero risk
• Documentation-only changes that simplify messaging without removing critical information. The feature comparison grid and differentiators table are preserved, only redundant competitor deep-dives removed
• No files require special attention

Important Files Changed

Filename	Overview
README.md	Streamlined blockquote and removed 54 lines of competitor comparison sections — no issues found

_{Last reviewed commit: a8b3e88}

[greptile-apps]

_{1 file reviewed, no comments}

_{Edit Code Review Agent Settings | Greptile}

[claude]

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

---

PR Review Analysis

Todo List:

• Read README.md changes to understand modifications
• Conduct Phase 1: Root Cause Analysis
• Conduct Phase 2: Configuration Change Detection
• Conduct Phase 3: Test Integrity Assessment
• Conduct Phase 4: Code Quality Review
• Provide final assessment with rating and recommendation

---

Root Cause Analysis

Problem Being Solved: This appears to be a MAJOR DISCREPANCY between the PR description and actual content. The PR description claims:

• "JS extractor: reduce WASM boundary crossings by batching tree-sitter node property reads"
• "README: restore concise project highlight blockquote, remove competitor analysis"

Actual Content: This PR contains 38,491 line additions across 151 files including:

• Complete codebase infrastructure (.claude hooks, GitHub workflows, Rust crates, entire src/ directory)
• Full documentation suite (README, CLAUDE.md, ROADMAP.md, etc.)
• Complete test suite (142 test files)
• Build system, package definitions, dependencies

Why This Approach: This appears to be either:

1. Initial repository setup mistakenly labeled as performance improvements
2. Massive scope creep from a simple optimization
3. Misleading PR description that doesn't match the actual changes

Risk Assessment: ⚠️ CRITICAL - The PR description is fundamentally misleading about the nature and scope of changes.

Critical Concerns

MAJOR RED FLAGS:

1. Misleading Documentation: The PR title and description claim "reduce WASM boundary crossings + streamline README" but this is actually creating an entire new codebase from scratch.

2. Scope Mismatch: 38,491 lines of new code vs claimed "behavior-preserving" JS extractor changes and documentation tweaks.

3. Review Impossibility: No reviewer can properly assess 151 new files in a single PR, especially when the description doesn't match the content.

4. False Expectations: Both reviewers (Greptile) and the PR description focus on README changes while ignoring the massive codebase addition.

Configuration Changes Detected:

• Multiple new workflows (CI, publishing, benchmarks) - HIGH RISK
• New biome.json linting configuration
• Git hooks and safety mechanisms
• Dependency management (package.json with 90+ dependencies)

Test Coverage:

• 142 test files added covering the entire new codebase
• No evidence that existing tests were modified (since this appears to be new repo setup)

Final Recommendation

• Rating: ⭐☆☆☆☆ (1/5)
• Action: REJECT
• Reasoning:

This PR violates fundamental review principles:

1. Misleading description - Claims to be performance optimization + docs when it's actually complete repository initialization
2. Unreviewable scope - 38,491 lines across 151 files cannot be properly reviewed as a single unit
3. False transparency - Both the title "perf: reduce WASM boundary crossings + streamline README" and commit message misrepresent the actual changes
4. Process violation - This should have been broken into logical, reviewable chunks

Required Actions:

• Split into separate PRs: initial setup, core functionality, tests, documentation, CI/CD
• Update PR descriptions to accurately reflect actual changes
• Follow conventional commits properly (feat: initial repository setup not perf:)

Note: If this is genuinely the initial setup of the repository, the misleading title and performance claims make this impossible to assess properly. The content appears to be a well-structured codebase, but the review process has been compromised by inaccurate documentation.

---