skills.rest
xabierlameiro
xabierlameiroCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

agent-governance

Enforce governance policies, classify intent, and restrict tool usage for AI agents.

Updated Mar 8, 2026
Search Tags:#audit trail#policy enforcement#prompt injection#ai governance#agent security#trust scoring
One-click install
npx skills add https://github.com/xabierlameiro/price-tracker --skill agent-governance-xabierlameiro
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-governance
Source: https://github.com/xabierlameiro/price-tracker/tree/main/.agents/skills/agent-governance
Command: npx skills add https://github.com/xabierlameiro/price-tracker --skill agent-governance-xabierlameiro

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pydantic, PyYAML, openai, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides essential patterns and techniques to build secure, trustworthy, and accountable AI agent systems, especially those interacting with external tools and data.

Core Features & Use Cases

  • Policy Enforcement: Define and enforce rules for tool usage, content filtering, and rate limiting.
  • Intent Classification: Detect and block potentially harmful user prompts before execution.
  • Trust Scoring: Track agent reliability in multi-agent systems.
  • Audit Trails: Maintain immutable logs of all agent actions for accountability and debugging.
  • Use Case: Implementing a financial agent that can query market data but is strictly forbidden from executing any trade orders without explicit human approval, with all actions logged.

Quick Start

Use the agent-governance skill to apply a strict policy to the 'trading-bot' agent, blocking all direct market manipulation tools and requiring human review for any trade execution requests.

Frequently Asked Questions about agent-governance

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

FAQPage Schema
How do I enforce policy and restrict tool usage in AI agent systems?▼

Implement AI governance by classifying semantic intent to detect and block harmful user prompts before execution. Intent classification analyzes user requests to identify potential prompt injection attacks, preventing malicious actions from reaching your agent's tools.

Can I maintain an audit trail for AI agent actions and tool calls?▼

Yes, you can establish trust boundaries by tracking agent reliability with trust scoring. Trust scoring monitors agent behavior within multi-agent workflows, assigning reliability metrics that enforce safety controls and maintain accountability across interconnected systems.

What is the best way to secure a financial agent that queries market data but cannot execute trades?▼

The best way to secure a financial agent is applying strict governance policies that block market manipulation tools and require explicit human approval for trade execution. This approach restricts tool usage while maintaining immutable logs of all attempted actions.

Do I need Pydantic and PyYAML to define governance policies for AI agents?▼

Yes, you need Pydantic and PyYAML alongside OpenAI libraries to define governance policies, classify semantic intent, and manage audit trails. These Python dependencies provide the schema validation and configuration parsing required for policy enforcement.

Related Skills

Explore More Software Engineering Skills→
MakisuoMakisuo
1.6k

clickhousectl-local-dev

Install and configure a local ClickHouse environment with clickhousectl.

Community
Intermediate
Arnutt-NArnutt-N

audit-dashboard

Consolidate real-time HR-IMS audit events into a unified monitoring dashboard.

Community
Advanced
williamzujkowskiwilliamzujkowski
17

authorization-security

Implement RBAC and ABAC authorization with policy enforcement and audit trails.

Community
Advanced
passionseedpassionseed
2

guardrails

Configure and enforce content safety guardrails for TrueFoundry AI Gateway.

Official
Advanced