agent-governance
Declarative policies, intent classification, and audit trails for controlling AI agent tool access and behavior.
What it does
- Agents with tool access: Any agent that calls external tools (APIs, databases, shell commands)
- Multi-agent systems: Agents delegating to other agents need trust boundaries
- Production deployments: Compliance, audit, and safety requirements
- Sensitive operations: Financial transactions, data access, infrastructure management
- searchdocuments
- querydatabase
- deleterecord
Requirements & configuration
- Multi-agent systems: Agents delegating to other agents need trust boundaries
- raise PermissionError(f"Tool '{toolname}' requires human approval")
- Deny or require explicit approval
- | Trust decay | Trust scores should decay over time — require ongoing good behavior |
Derived from the skill's own SKILL.md documentation · extracted 2026-07-23
Declarative policies, intent classification, and audit trails for controlling AI agent tool access and behavior.
Source
Repository: https://github.com/github/awesome-copilot
agent-governance FAQ
What does the agent-governance skill do?
| Patterns and techniques for adding governance, safety, and trust controls to AI agent systems. Use this skill when: - Building AI agents that call external tools (APIs, databases, file systems) - Implementing policy-based access controls for agent tool usage - Adding semantic intent classification to detect dangerous prompts - Creating trust scoring systems for multi-agent workflows - Building audit trails for agent actions and decisions - Enforcing rate limits, content filters, or tool restrictions on agents - Working with any agent framework (PydanticAI, CrewAI, OpenAI Agents, LangChain, AutoGen) Agents with tool access: Any agent that calls external tools (APIs, databases, shell commands) Multi-agent systems: Agents delegating to other agents need trust boundaries
What does agent-governance require?
Multi-agent systems: Agents delegating to other agents need trust boundaries raise PermissionError(f"Tool '{toolname}' requires human approval") Deny or require explicit approval | Trust decay | Trust scores should decay over time — require ongoing good behavior |
How do I install agent-governance?
Run: npx -y skills add https://github.com/github/awesome-copilot --skill agent-governance --agent claude-code — the source lives at github.com/github/awesome-copilot.
Maintain agent-governance?
Let people know it's listed here — add the badge (live metrics, light/dark aware) or a plain link to your README or docs.
[](https://getagentictools.com/skills/github-awesome-copilot-agent-governance?ref=badge) npx agentictools info skills/github-awesome-copilot-agent-governance The second line is the CLI lookup for this page — handy in READMEs and docs.
Related skills
readme-blueprint-generator Automated README generation by analyzing project documentation structure and metadata files.
git-flow-branch-creator Analyzes git changes and creates semantic Git Flow branches automatically based on change type.
azure-role-selector Guidance for selecting least-privilege Azure roles matching specific permission requirements.
copilot-sdk Programmatic agent runtime for embedding Copilot's agentic workflows in applications across Python, TypeScript, Go, and .NET.
context-map Analyze codebases and map task-relevant files before implementing changes.
write-coding-standards-from-file Analyze existing code files to automatically generate a project coding standards document.