← ClaudeAtlas

kraitlisted

AI-first security auditor for Solidity smart contracts. 4-phase pipeline (recon → detection → state analysis → verification) with 101 heuristics, 15 detection modules, 8 kill gates, and a 6-field methodology audit trail per finding (v8.1). Tested at 100% precision across 50 blind shadow audits.
ZealynxSecurity/krait · ★ 21 · AI & Automation · score 77
Install: claude install-skill ZealynxSecurity/krait
# Krait — AI Security Auditor Krait is a structured audit methodology for Solidity smart contracts, encoded as Claude Code skills. It runs a 4-phase pipeline with multi-mindset analysis and strict verification gates. ## How It Works When invoked via `/krait`, a preflight readiness check runs first, then the 4-phase pipeline runs sequentially: - **Preflight** (`preflight/instructions.md`): Hard checks (forge / bash / jq / `.sol` files) before any work. Same skill is invoked in *report* mode by `/krait-init` and in *gate* mode by `/krait`. 1. **Phase 0 — Recon** (`recon/instructions.md`): Architecture mapping, deterministic file risk scoring, module selection 2. **Phase 1 — Detection** (`detector/instructions.md`): 3-pass analysis with 4 parallel lenses × 4 mindsets, 101 heuristics, activated detection modules 3. **Phase 1b — Rescan** (`detector/rescan.md`): Second broad pass with pass-1 findings as an exclusion list; counters attention saturation. Self-skips when pass 1 found nothing above Info 4. **Phase 1c — Per-Contract** (`detector/per-contract.md`): One agent per inheritance cluster at maximum depth; counters attention dilution 5. **Phase 2 — State Analysis** (`state-auditor/instructions.md`): Coupled state pair analysis, mutation matrix, masking code detection 6. **Phase 3 — Verification** (`critic/instructions.md`): 8 automatic kill gates + Impact Premise (harm, not mechanism); concrete exploit trace required for every H/M finding 7. **Phase 3b — Review** (`reviewer