amp-permissions

Community

Fine-grained control over tool permissions.

Authorthurstonsand
Version1.0.0
Installs0

System Documentation

What problem does it solve?

Amp’s permission system controls every tool invocation by evaluating a prioritized list of rules before execution. It provides consistent policy enforcement across main threaded operations and subagents, enabling allow, reject, ask, or delegate actions for diverse tools. This enables safe, auditable automation in AI-assisted workflows by centralizing decision logic and providing testable configurations.

Core Features & Use Cases

  • Rule-based decisions: allow, reject, ask, or delegate tool calls with a single, ordered policy.
  • Context-aware enforcement: applies to both the main thread and subagent tool executions to prevent leakage or unsafe actions.
  • Testable policies: supports dry-runs and testing utilities to validate rules before deployment.

Quick Start

Define your first permission rule and test its effect on a tool invocation.

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: amp-permissions
Download link: https://github.com/thurstonsand/ansiblonomicon/archive/main.zip#amp-permissions

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 223,000+ vetted skills library on demand.