⬢github TypeScript · 1.1K ★ +288 since we first saw it · pushed 14 h ago · MIT
qybaihe/bla
🛡️ BLA (Blind LLM Agents) — Adversarial defense shield that protects binaries, Electron apps, JS bundles, Python, and WebAssembly against AI reverse-engineering agents like REA. Poison -> Exhaust -> Hallucinate.
BLA (Blind LLM Agents) is a defensive TypeScript tool that hardens your shipped software—binaries, Electron apps (.asar), JS bundles, Python, WebAssembly—against AI reverse-engineering agents. It poisons artifacts with decoy canaries and control flow, exhausts analyzers' resource budgets, and leaves them hallucinating, all while keeping runtime behavior unchanged. It runs via CLI or MCP and includes a benchmark.
Why now: Trending on GitHub's most-starred new repos list, riding current concern over LLM-powered reverse-engineering agents and offering an empirical counter-REA defense.
Who it is for: Developers shipping compiled or bundled software who want to protect their own artifacts from AI-driven reverse engineering.
adversarial-aiclaude-codecursorelectron-securityllm-securitymcpmcp-serverobfuscationreverse-engineeringwebassembly
Stars over our 30 snapshots: 768 to 1.1K, since 7 h ago.
Where people talked about it
- ⬢github new repos, most starred 13 min ago
API: https://socialmediatrends-api.osmike.com/v1/repos/qybaihe/bla