# Security Expert Engine

> Use when running the monthly security EXPERTS knowledge loop — reading, diffing, and updating the security expert digest without fabrication.

- Skill: `metinduraktr-44/security-expert-engine-2` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add metinduraktr-44/security-expert-engine-2`
- Raw SKILL.md: https://api.skillmd.com/api/skills/metinduraktr-44/security-expert-engine-2/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Security
- Author: metinduraktr-44 (https://skillmd.com/u/metinduraktr-44)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/metinduraktr-44/security-expert-engine-2

---


# GUARDRAIL AKTİF — savunma-only, secret-redakte, exploit-yok

# Security Expert Engine

## Instructions
1. Run the monthly loop in `EXPERTS/README.md`: READ `EXPERTS/SECURITY_DIGEST.md` → DELTA (dated, only if research enabled) → DIFF → WRITE → DIGEST.
2. Seeds (Schneier, Boneh, Moussouris, Kaminsky[historical 2021], …) are real public figures — mark bios/URLs `araştırılacak / URL doğrulanmalı`.
3. Track standards drift (NIST CSF 2.0, 800-53 Rev.5, ISO 27001:2022, CIS v8.1, OWASP ASVS 5.0.0, PQC, SLSA, ZTMM). No fabrication, no secrets.

## References
- `references/OUTLINE.md` — depth outline (filled in phases).

## Note
Full ~20k-char content is produced later in phases. This is the discoverable skeleton.