What is the primary framework used by AI Risk Mapper?

The skill primarily utilizes the Coalition for Secure AI (CoSAI) Risk Map framework for identifying and mitigating security risks in AI systems.

Which compliance frameworks are supported?

AI Risk Mapper aligns with MITRE ATLAS, NIST AI RMF, OWASP Top 10 for LLM, STRIDE, and ISO 22989.

Can I use AI Risk Mapper without an internet connection?

Yes, it supports an --offline flag which utilizes bundled schemas, allowing for risk assessments in air-gapped or restricted network environments.

What kind of targets can I analyze?

You can analyze local codebases, file paths, architecture documents, or provide brief system descriptions for the orchestrator to evaluate.

AI Risk Mapper

Name: AI Risk Mapper
Author: totallyGreg

bytotallyGreg

•

安全与测试

Identifies, analyzes, and mitigates security risks in AI systems and LLM applications using the CoSAI Risk Map framework.

AI Risk Mapper is a specialized security tool for Claude Code designed to help developers and security engineers harden AI infrastructure and machine learning pipelines. By leveraging the Coalition for Secure AI (CoSAI) Risk Map, the skill automates the identification of vulnerabilities, maps threats to industry standards like MITRE ATLAS and NIST AI RMF, and generates comprehensive risk assessment reports. Whether you are building RAG applications or managing model-serving infrastructure, this skill provides the necessary tooling to perform deep threat modeling, persona-based risk profiling, and control gap analysis to ensure your AI systems are secure by design.

主要功能

011 GitHub stars

02Gap analysis tools to evaluate current security control coverage

03Automated AI security risk assessments and comprehensive CoSAI reporting

04Interactive CLI for risk searching, control identification, and persona profiling

05Offline mode support for secure environments using bundled schemas

06Mapping of threats to MITRE ATLAS, NIST AI RMF, and OWASP Top 10 for LLM

使用场景

01Mapping specific LLM vulnerabilities to MITRE ATLAS techniques for compliance reporting

02Performing threat modeling for different personas like Model Providers or Application Developers

03Conducting a security audit on an AI chatbot codebase to identify prompt injection risks

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add totallygreg/claude-mp ai-risk-mapper

For use in Claude.ai and ChatGPT

主要功能

011 GitHub stars

02Gap analysis tools to evaluate current security control coverage

03Automated AI security risk assessments and comprehensive CoSAI reporting

04Interactive CLI for risk searching, control identification, and persona profiling

05Offline mode support for secure environments using bundled schemas

06Mapping of threats to MITRE ATLAS, NIST AI RMF, and OWASP Top 10 for LLM

使用场景

01Mapping specific LLM vulnerabilities to MITRE ATLAS techniques for compliance reporting

02Performing threat modeling for different personas like Model Providers or Application Developers

03Conducting a security audit on an AI chatbot codebase to identify prompt injection risks