Helios – 3 Claude agents (Red vs. Blue) hack and patch your codebase
Red vs Blue AI agents battling over your code beats static scanning.
Variant analysis, open-sourced. Feed a CVE patch, find every structural twin across your codebase. 6-stage pipeline: patch ingestion, LLM signature extraction, embedding search, Joern CPG slicing, LLM feasibility reasoning, SARIF reporting.
Six-stage pipeline with Joern CPG and LLM reasoning beats simple pattern matching.
Security engineers, DevSecOps teams
Semgrep · CodeQL · GitHub Advanced Security
Red vs Blue AI agents battling over your code beats static scanning.
Catches AI code debt deterministically with AST analysis when linters only check style.
Runtime behavior blocking for zero-days, but does it beat Falco or eBPF-based tools?
Commit-time indexing beats runtime scanning, but Cursor already indexes your codebase.
AST-based contracts beat raw code for AI, but codebase analyzers are increasingly crowded.
Yet another secret scanner, but this one's a single Python file.