Skip to content

Latest commit

 

History

History
37 lines (30 loc) · 1.59 KB

File metadata and controls

37 lines (30 loc) · 1.59 KB

Status

v0.1.0 — Initial Release (2026-04-17)

Complete

  • 6-stage pipeline: Ingest -> Distill -> Retrieve -> Slice -> Reason -> Report
  • 4 LLM backends: Ollama (local), Groq (free), Anthropic, OpenAI — all with rate limit retry
  • 2 embedding backends: Ollama (local), Voyage AI
  • 8 languages: C, C++, Java, Python, JavaScript, TypeScript, Go, Rust
  • tree-sitter function extraction across all supported languages
  • Joern CPG analysis with auto-download of Java + Joern
  • SARIF 2.1.0 output for CI integration
  • 6 rulepacks: C memory safety, Java deserialization, Python injection, Go concurrency, Rust safety, ML framework security
  • 15 real CVEs in benchmark corpus across 8 vulnerability families
  • Embedding cache — re-runs skip the embedding step entirely
  • NVD API caching with rate limit retry
  • Custom benchmarks via --corpus flag
  • 44 tests including E2E pipeline test with mock LLM
  • Published to crates.iocargo install naptrace
  • Animated CLI banner (figlet slant font)
  • Cross-platform — macOS, Linux, Windows (partial)

Tested against

Codebase Language Functions Accuracy
Redis (177K lines) C 5,335 Found real potential overflow variant
Flask Python ~200 0 false positives
Gson Java ~300 0 false positives, identified anti-deser guards

What's next

  • More benchmark corpus entries with verified ground truth
  • Diff-only scanning mode (PR integration)
  • Community rulepack contributions
  • Performance optimization for large codebases