# Ticket Risk Policy

> Use when implementing or reviewing risk assessment, human review routing, complaint handling, compensation handling, legal risk, privacy risk, or account security risk.

- Skill: `lennney/ticket-risk-policy` (Agent Skill)
- Install (CLI): `npx skillmds@latest add lennney/ticket-risk-policy`
- Raw SKILL.md: https://api.skillmd.com/api/skills/lennney/ticket-risk-policy/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Security
- Author: lennney (https://skillmd.com/u/lennney)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/lennney/ticket-risk-policy

---


# Ticket Risk Policy

Risk flags:
- complaint_risk
- compensation_risk
- legal_risk
- privacy_risk
- account_security_risk
- policy_conflict
- insufficient_evidence
- low_confidence

Hard rules:
1. Complaint, compensation, legal, privacy, and account security cases must enter human review.
2. Insufficient evidence must not produce a confident final reply.
3. Low classification confidence must enter human review.
4. Policy conflict must enter human review.
5. The system is a customer support assistant, not a fully automated support agent.