tool-calling-approval-reentry
CommunityEnsure destructive AI tools always pause for human approval before running.
Software Engineering#AI safety#human-in-the-loop#safety checks#tool control#AI tool calls#multi-turn conversation
Authormarvelousempire
Version1.0.0
Installs0
System Documentation
What problem does it solve?
Prevents AI agents from running destructive tools without human oversight, ensuring all destructive actions are reviewed and approved.
Core Features & Use Cases
- Human-in-the-Loop Approval: Ensures that all destructive tool calls are reviewed by a human before execution.
- Multi-Turn Loop: Handles complex approval processes that may require multiple interactions.
- Hard Cap on Rounds: Protects against runaway tool loops by enforcing a maximum number of rounds.
- Cross-Platform Compatibility: Functions across Anthropic Messages API and OpenAI function-calling.
Quick Start
Invoke "tool-calling-approval-reentry" to pause destructive AI tool calls and require human approval.
Dependency Matrix
Required Modules
None requiredComponents
scripts
💻 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: tool-calling-approval-reentry Download link: https://github.com/marvelousempire/ai-skills-library/archive/main.zip#tool-calling-approval-reentry Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
Agent Skills Search Helper
Install a tiny helper to your Agent, search and equip skill from 620,000+ vetted skills library on demand.