aigis

Classify AI systems into governance areas and run deterministic codebase verification.

5|Updated Mar 26, 2026
One-click install
npx skills add https://github.com/NayanKanaparthi/aigis --skill aigis
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: aigis
Source: https://github.com/NayanKanaparthi/aigis/tree/main
Command: npx skills add https://github.com/NayanKanaparthi/aigis --skill aigis

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Aigis eliminates the risk that coding agents ship AI features without proper governance by turning governance frameworks into executable agent instructions and deterministic verification steps.

Core Features & Use Cases

  • Classify governance scope from a description: Detects relevant governance traits for your AI system using a curated, checksum-guarded resolver map.
  • Fetch step-by-step governance procedures: Retrieves area-specific implementation patterns (input validation, PII handling, prompt security, audit logging, and more) and workflow layouts when needed.
  • Verify deterministically against your codebase: Runs regex-based scanners and targeted recognizers to surface PASS/FAIL/OVERCLAIM signals with evidence, then compiles audit-ready reports.
  • Generate audit-ready compliance documentation: Produces structured reports with cross-framework citations (OWASP, NIST AI RMF, ISO/IEC 42001, and EU AI Act when opted into).

Quick Start

Use Aigis to build a governance brief for an EU AI Act-ready chatbot by running: aigis build "build an EU AI Act compliant customer support chatbot for an e-commerce site"

Frequently Asked Questions about aigis

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate EU AI Act compliance verification for my LLM project?

Automate EU AI Act compliance verification by classifying your AI system against governance traits, fetching implementation procedures, and running regex-based scanners over your codebase to generate audit-ready reports. This verifies features like PII handling and prompt security deterministically.

What is deterministic AI governance verification for coding agents?

Deterministic AI governance verification uses regex-based scanners and targeted recognizers to evaluate your codebase against framework requirements. It surfaces PASS, FAIL, or OVERCLAIM signals with evidence, ensuring coding agents ship compliant AI features.

Can I use this to audit existing AI projects for OWASP and NIST AI RMF compliance?

Yes, you can audit existing AI projects by running targeted scanners over the codebase to verify compliance against OWASP, NIST AI RMF, ISO/IEC 42001, and EU AI Act frameworks. It compiles structured reports with cross-framework citations and evidence.

How do I build governance procedures for an AI customer support chatbot?

Build governance procedures by providing a system description to detect relevant governance traits. The resolver fetches step-by-step implementation patterns for input validation, PII handling, prompt security, and audit logging tailored to your chatbot.

Does this AI governance tool require specific dependencies to run codebase verification?

No external dependencies are required to run codebase verification. The tool operates autonomously using a resolver-driven trait pipeline and on-demand skill fetching to scan your target codebase and compile compliance documentation.

What are the limitations of regex-based compliance scanners for AI systems?

Regex-based compliance scanners provide deterministic verification by matching specific code patterns against framework requirements. They surface concrete PASS, FAIL, or OVERCLAIM signals but are limited to recognizable code structures rather than semantic logic.