Skills
.Work. You
Rest

ai-agent-guardrails

Community

Guardrails for AI agents handling real money.

Finance & Accounting#rate-limiting#guardrails#identity-verification#ai-safety#prompt-hardening#financial-transactions#two-call-llm
Authorazep-ninja
Version1.0.0
Installs0

System Documentation

What problem does it solve?

AI agents handling real-money transactions face prompt injection, hallucinations, identity confusion, and abuse risks. This Skill defines a reusable, framework-agnostic guardrail pattern that enforces safety across three layers of defense and provides deterministic checks between LLM steps.

Core Features & Use Cases

  • Two-call LLM architecture to reduce hallucinations and misclassification
  • Deterministic validation for operation completeness, identity verification, and rate limiting
  • Execution safety with deduplication, platform restrictions, and pre-execution safeguards
  • Open-source guardrail patterns with production-tested usage across multiple networks

Quick Start

Integrate the two-call LLM pattern with deterministic validation to safely guard financial operations in your agent pipeline.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: ai-agent-guardrails
Download link: https://github.com/azep-ninja/ai-agent-guardrails/archive/main.zip#ai-agent-guardrails

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.