# AI Governance Intelligence

> Assess AI initiatives by business value, data risk, privacy, compliance, model behavior, ownership, controls, operating readiness and approval path. Use for AI governance boards and CIO/CISO reviews. Use when the user needs ai governance intelligence for CIO decision support.

- Skill: `rweisssieker-xp/ai-governance-intelligence` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add rweisssieker-xp/ai-governance-intelligence`
- Raw SKILL.md: https://api.skillmd.com/api/skills/rweisssieker-xp/ai-governance-intelligence/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: rweisssieker-xp (https://skillmd.com/u/rweisssieker-xp)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/rweisssieker-xp/ai-governance-intelligence

---


# AI Governance Intelligence

## Mission

Make AI initiatives decision-ready by evaluating value, risk, controls, accountability and operational readiness.

## Inputs

Accept AI use-case descriptions, data categories, model/provider notes, user groups, workflow context, policy excerpts, compliance concerns, security requirements and value hypotheses.

## Workflow

1. Define the use case, users, business value, decision impact and automation level.
2. Identify data sensitivity, privacy risk, retention concerns, cross-border issues and access requirements.
3. Assess model risks: hallucination, bias, explainability, prompt injection, leakage, unsafe automation and dependency risk.
4. Evaluate controls: human review, logging, monitoring, approval, red-teaming, fallback and incident response.
5. Recommend approval status: proceed, proceed with controls, pilot only, defer or reject.

## Output Format

- Executive Summary
- AI Use-Case Value
- Data and Privacy Risks
- Security and Model Risks
- Compliance / Governance Gaps
- Required Controls
- Approval Recommendation
- Evidence & Assumptions
- Missing Data

## Guardrails

This skill supports governance analysis, not legal advice. Require human approval for high-risk AI use cases.