CSlawyer1985/legal-skillhub / skills/ai-compliance
ai-compliance
VerifiedAI compliance analysis for EU AI Act, ISO 42001, NIST AI RMF, GDPR, OECD, financial services regulations (SEC, FCA, FINRA, DORA, MiFID II), and other frameworks. Use when asked to generate a compliance checklist for an AI tool or use case, determine if a risk assessment is required, score an AI tool's risk level, identify where an AI tool or use case could run afoul of regulatory requirements, perform a gap analysis, recommend remediation steps, assess a vendor, draft an acceptable use policy, map training requirements, or review jurisdiction-specific AI rules. Triggers on phrases like "compliance checklist", "risk assessment", "risk score", "does this comply", "EU AI Act", "ISO 42001", "NIST AI RMF", "AI governance", "gap analysis", "vendor assessment", "acceptable use policy", "MNPI", "AI training requirements", or any request to evaluate an AI tool or use case for regulatory compliance or risk.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 4.6 KB |
| _meta.json | config | 132 B |
| references/ai-inventory.md | markdown | 2.7 KB |
| references/aup-template.md | markdown | 3.7 KB |
| references/checklist-templates.md | markdown | 4.1 KB |
| references/data-classification.md | markdown | 3.7 KB |
| references/eu-ai-act.md | markdown | 2.6 KB |
| references/finserv-regulations.md | markdown | 4.7 KB |
| references/incident-response.md | markdown | 4.9 KB |
| references/iso-42001.md | markdown | 3.4 KB |
| references/iso27001-alignment.md | markdown | 4.4 KB |
| references/jurisdiction-map.md | markdown | 4.3 KB |
| references/nist-ai-rmf.md | markdown | 4.6 KB |
| references/other-frameworks.md | markdown | 2.6 KB |
| references/remediation-playbooks.md | markdown | 6.6 KB |
| references/risk-scoring.md | markdown | 4.4 KB |
| references/training-requirements.md | markdown | 3.3 KB |
| references/vendor-assessment.md | markdown | 3.5 KB |
references/nist-ai-rmf.md
View on GitHub# NIST AI Risk Management Framework (AI RMF 1.0)
## Overview
Voluntary framework published January 2023. Helps organizations manage risks from AI systems across the lifecycle. Organized around four core functions: GOVERN, MAP, MEASURE, MANAGE.
## Core Functions
### GOVERN
Establishes organizational practices, culture, and accountability for AI risk.
**Key subcategories:**
- GOVERN 1: Policies and processes for AI risk management are documented and communicated
- GOVERN 1.1: AI risk management framed in policies, processes, procedures, practices
- GOVERN 1.2: Accountability for AI risks assigned
- GOVERN 1.3: Organizational teams understand their AI risk responsibilities
- GOVERN 1.4: Organizational teams are committed to transparency
- GOVERN 1.5: Organizational risk tolerance for AI is established
- GOVERN 1.6: AI risk management policies include diversity, equity, inclusion considerations
- GOVERN 1.7: Processes for AI risk management are established and maintained
- GOVERN 2: AI risk management workforce/culture cultivated
- GOVERN 3: Feedback processes for AI risk management are in place
- GOVERN 4: Organizational teams are committed to risk management
- GOVERN 5: Policies and processes address AI risks and benefits to third parties
- GOVERN 6: Risk management policies address AI risks from third-party entities
### MAP
Identifies and categorizes AI risks in context.
**Key subcategories:**
- MAP 1: Context established and understood
- MAP 1.1: Intended purpose, benefits, context, and risks documented
- MAP 1.5: Organizational risk tolerances are communicated for AI deployment
- MAP 1.6: AI system categorization consistent with risk management priorities
- MAP 2: Scientific understanding of AI risks is examined
- MAP 2.1: Scientific evidence for AI risks is reviewed
- MAP 2.2: Practices for scientific rigor in AI risk assessment
- MAP 3: AI system risks are identified
- MAP 3.1: Potential impacts to individuals and communities assessed
- MAP 3.2: Data collected and used for training/operation is known
- MAP 3.5: Likelihood of AI risks is estimated
- MAP 4: Risks are prioritized
- MAP 5: Likelihood/impact of AI risks tracked over time
### MEASURE
Analyzes and assesses AI risks using quantitative and qualitative methods.
**Key subcategories:**
- MEASURE 1: Methods for measuring risk are established
- MEASURE 2: AI risks are evaluated
- MEASURE 2.1: Test sets for bias, fairness, accuracy documented
- MEASURE 2.2: Evaluations are human- and AI-relevant
- MEASURE 2.3: AI system performance is evaluated
- MEASURE 2.5: Privacy risks evaluated
- MEASURE 2.6: Explainability/interpretability evaluated
- MEASURE 2.7: AI system security/resilience evaluated
- MEASURE 2.8: Risks from AI supply chain evaluated
- MEASURE 2.9: Impacts of system errors/failure assessed
- MEASURE 3: Risk assessment results are documented
- MEASURE 4: Feedback informs measurement practices
### MANAGE
Prioritizes and addresses AI risks.
**Key subcategories:**
- MANAGE 1: Risks are prioritized and responded to
- MANAGE 1.1: Risks based on impact/likelihood are responded to
- MANAGE 1.3: Response to AI risks includes contingency plans
- MANAGE 2: Risk strategies are applied
- MANAGE 2.1: Plans for treating identified risks are implemented
- MANAGE 2.2: Mechanisms for feedback about AI risk treatments
- MANAGE 2.4: Residual risks are documented
- MANAGE 3: Risks are tracked
- MANAGE 3.1: AI risks to be monitored are identified
- MANAGE 3.2: Responses to AI risks are implemented
- MANAGE 4: Risks from past incidents are used to improve
- MANAGE 4.1: Post-incident lessons are documented
## TRUSTWORTHY AI CHARACTERISTICS (AI RMF Core)
NIST defines trustworthy AI across these properties:
1. **Accountable** — clear ownership and responsibility
2. **Explainable/Interpretable** — decisions can be understood
3. **Fair** (bias managed) — does not discriminate unfairly
4. **Privacy-Enhanced** — protects personal data
5. **Reliable/Robust** — performs consistently, handles adversarial inputs
6. **Safe** — does not cause harm
7. **Secure/Resilient** — resistant to attacks
8. **Transparent** — operations are open to scrutiny
## Key Compliance Questions
1. Is there an AI risk management policy and process?
2. Are AI use cases inventoried and categorized by risk?
3. Is the intended purpose of each AI system documented?
4. Are bias, fairness, and accuracy evaluations conducted?
5. Is human oversight defined?
6. Are third-party AI vendors assessed for risk?
7. Are incidents tracked and used to improve the program?
8. Is AI risk tolerance defined and communicated?
9. Are privacy impacts assessed?
10. Is explainability/transparency evaluated and documented?
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| receiving-code-reviewobra/superpowersUse when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| requesting-code-reviewobra/superpowersUse when completing tasks, implementing major features, or before merging to verify work meets requirements | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| using-git-worktreesobra/superpowersUse when starting feature work that needs isolation from current workspace or before executing implementation plans - ensures an isolated workspace exists via native tools or git worktree fallback | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| verification-before-completionobra/superpowersUse when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |