Aguardic logoAguardic

AI Agent Governance

by AguardicOfficial·v1.0.0

Govern autonomous AI agent actions before they execute. Enforce permission boundaries, require human approval for consequential actions, detect scope violations, and maintain full audit trails for every agent decision.

About This Policy Template

Comprehensive governance pack for autonomous AI agents. Your AI agents don't just generate text — they send emails, modify documents, execute code, make API calls, and take actions across your systems. One unauthorized action can expose data, violate compliance requirements, or commit your organization to something it didn't approve. This pack enforces permission boundaries, detects prompt injection targeting agents, requires human approval for consequential actions, flags cross-system data transfers, and ensures every agent action has a complete audit trail. Built in response to the NIST AI Agent Standards Initiative and designed for organizations deploying agentic AI in production.

Policy Rules(10)

Critical Severity

(3)

Agent Action Outside Authorized Scope

Detect when an AI agent attempts to perform actions outside its defined permission boundaries — accessing systems, APIs, or data it should not touch.

AI

Agent Privilege Escalation Attempt

Detect when an AI agent attempts to gain elevated permissions, modify access controls, create new credentials, or bypass authorization mechanisms.

AI

Untrusted Data Influencing Agent Actions

Detect patterns indicating that untrusted external data (emails, web content, user-submitted documents) may contain injected instructions that could manipulate agent behavior.

AI

High Severity

(4)

Consequential Action Without Human Approval

Flag agent actions that have significant real-world impact (financial transactions, data deletion, external communications, production deployments) taken without documented human approval.

AI

External Communication Without AI Disclosure

Detect when an AI agent sends external communications without disclosing that the content was generated or sent by an AI system.

AI

Sensitive Data in Agent Context Window

Detect when an AI agent's context window or action payload contains sensitive data (PII, credentials, financial data) that the agent should not be processing.

Rule

Unauthorized Cross-System Data Transfer

Detect when an AI agent moves or copies data between systems without authorization — such as copying customer data from a CRM to a messaging platform.

AI

Medium Severity

(3)

Agent Action Without Sufficient Audit Trail

Detect when an AI agent action lacks the minimum metadata required for a complete audit trail — including timestamp, action type, target system, agent identity, and outcome.

AI

Agent Operating Outside Approved Schedule

Detect when an AI agent takes actions outside its approved operating windows or frequency limits.

AI

Multi-Step Action Chain Without Checkpoint

Flag when an AI agent executes a sequence of multiple consequential actions without intermediate checkpoints, approval gates, or pause points.

AI

Enforcement by Integration

What happens when a violation is detected, based on the enforcement mode and integration type.

IntegrationBlockApprovalWarnMonitor
Version Control
GitHub, GitLab, Bitbucket
Fail check run / merge request statusPending check run — held for reviewNeutral check run / comment on PRPass check run (silent)
Email — Gmail
Gmail
Quarantine label; + violation label (outbound)Quarantine label — held for reviewAdd warning labelLog only
Email — Outlook
Outlook
Move to quarantine folder; + flag (outbound)Move to quarantine — held for reviewFlag + categorizeLog only
Messaging
Slack, Teams
Post violation warning in channelPost 'held for review' warningPost warning in channelLog only
Storage
Google Drive, Dropbox, OneDrive
Move file to quarantine folderQuarantine file — held for reviewLog onlyLog only
AI Proxy
OpenAI, Anthropic, Gemini, MCP, Agent
Block request (return 403)Hold request — return review IDAllow request + audit trailLog only
API
REST API
Return BLOCK outcome (client decides)Return APPROVAL_REQUIRED + poll URLReturn WARN outcomeLog only

Version History

1 version published

v1.0.0Active2/23/2026

Initial release

Federal / enterprise questionnaire?

Answer NIST AI RMF questions with function-level controls Aguardic enforces

Try the tool

Ready to Install AI Agent Governance?

Get started with pre-built governance policies in minutes.