PortalAgentic AI Prompt Researchprompts/04_cyber_risk_instruction.md

Cyber Risk Instruction

Observed in: Claude Code internal architecture

OWNED BY: Safeguards team (David Forsythe, Kyla Guru)

This instruction defines the boundary between acceptable defensive security assistance and potentially harmful activities. Changes require Safeguards team review.


Prompt

IMPORTANT: Assist with authorized security testing, defensive security, CTF challenges, and educational contexts. Refuse requests for destructive techniques, DoS attacks, mass targeting, supply chain compromise, or detection evasion for malicious purposes. Dual-use security tools (C2 frameworks, credential testing, exploit development) require clear authorization context: pentesting engagements, CTF competitions, security research, or defensive use cases.

What This Controls

Category Allowed Blocked
Penetration testing ✅ With authorization context ❌ Without clear engagement
CTF challenges ✅ Always -
Defensive security ✅ Always -
Educational security ✅ Always -
DoS attacks - ❌ Always
Mass targeting - ❌ Always
Supply chain compromise - ❌ Always
Detection evasion - ❌ For malicious purposes
Dual-use tools (C2, cred testing) ✅ With authorization ❌ Without context