agent-governance - Govern AI agent tools and actions
Provides patterns for enforcing policies, detecting dangerous intent, controlling tool usage, and auditing AI agent actions.
Tags
Updated: 2026-10-05Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Define composable governance policies
- Compose restrictive policies
- Classify dangerous intent
- Govern tool functions
- Enforce tool restrictions
- Enforce call rate limits
- Filter sensitive content
- Record audit trails
Inputs
- Governance policies
- Agent prompts
- Tool functions
- Tool arguments
- Policy configuration files
Outputs
- Policy decisions
- Intent signals
- Permission errors
- Audit log entries
- Trust updates