ai-security-verification
Comprehensive AI security verification using OWASP AI Security Verification Standard (AISVS) framework. Provides structured checklist to verify security and ethical considerations across 13 categories of AI-driven applications, from training data governance to human oversight.
57
36%
Does it follow best practices?
Impact
98%
1.06xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./.claude/skills/ai-security-verification/SKILL.mdEvaluation results
96%
3%Contract Intelligence Platform Security Assessment
Full AISVS compliance report for RAG legal AI platform
AISVS Compliance Assessment
50%
100%
Security Control Evaluation
100%
75%
Gap Analysis
62%
75%
Risk-Based Prioritization
100%
100%
Compliance Roadmap
100%
100%
Verification Evidence
100%
100%
Category breadth
100%
100%
Training Data Governance
100%
100%
User Input Validation
100%
100%
Finding format compliance
100%
100%
Monitoring & Logging
100%
100%
Human Oversight
100%
100%
100%
1%Autonomous Email Agent Security Review
Agentic AI autonomous email agent security review
Agentic Action Security
100%
100%
Prompt injection risk
100%
100%
Adversarial Robustness
87%
100%
Output Control & Safety
100%
100%
Supply Chain Security
100%
100%
AISVS Compliance Assessment
100%
100%
Security Control Evaluation
100%
100%
Gap Analysis
100%
100%
Risk-Based Prioritization
100%
100%
Compliance Roadmap
100%
100%
OWASP reference
100%
100%
100%
14%Medical AI Compliance and Security Assessment
Medical AI privacy, governance, and human oversight assessment
Training Data Governance
100%
100%
Privacy Protection
100%
100%
Human Oversight
100%
100%
Model Lifecycle Management
100%
100%
Supply Chain Security
100%
100%
Compliance Roadmap
100%
100%
Verification Evidence
100%
100%
Gap Analysis
100%
100%
Risk-Based Prioritization
100%
100%
OWASP references
0%
100%
- Repository
- OWASP/secure-agent-playbook
- Commit
3f4fcb6
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.