loop-maker

Scaffold autonomous agent loops with verifiers, state files, and human gates.

67|12|Updated Jun 24, 2026

Search Tags:#automation #workflow #scaffolding #verification #autonomous #agentic-loops

Ready to run

npx skills add https://github.com/EricTechPro/loop-maker --skill loop-maker

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: loop-maker
Source: https://github.com/EricTechPro/loop-maker/tree/main
Command: npx skills add https://github.com/EricTechPro/loop-maker --skill loop-maker

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill prevents common agentic loop failures such as runaway execution, state leakage, and the lack of a reliable, separate verification mechanism.

Core Features & Use Cases

7-Question Wizard: Guides you through defining goals, triggers, discovery, actions, verification, state, and human gates.
Durable Scaffolding: Generates a complete, ready-to-run loop folder including a verifier script, state file, and human-gate documentation.
Use Case: Automate a recurring data-sync job or a monitoring agent by ensuring it has a checkable exit predicate and a hard budget, preventing infinite loops and resource exhaustion.

Quick Start

Invoke the loop-maker skill to begin the interactive wizard and scaffold your autonomous agent loop.

Frequently Asked Questions about loop-maker

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I prevent infinite loops and runaway execution when building autonomous agents?▼

Prevent infinite loops in autonomous agents by enforcing a 7-question blueprint that mandates separate verifier scripts, external state files, and hard human gates. This scaffolding ensures cold-start resilience and strict budget control to stop resource exhaustion.

What is the best way to scaffold a self-verifying monitoring agent for recurring tasks?▼

The best way to scaffold a self-verifying monitoring agent is using a wizard that generates a ready-to-run loop folder. This includes a dedicated verifier script, an external state file, and human-gate documentation to ensure reliable, checkable exit predicates.

Does this agentic workflow scaffolding work with Claude Code and Codex?▼

Yes, the agentic workflow scaffolding is host-neutral and portable across Claude Code, Codex, Hermes, and OpenClaw. It generates workflows that satisfy the requirement for portable automation without relying on environment-specific dependencies.

How do I add human gates to an automated data-sync pipeline?▼

Add human gates to an automated data-sync pipeline by defining them during the scaffolding process. The generated loop folder includes explicit human-gate documentation, ensuring that recurring automated tasks pause for manual approval before proceeding.

Why does my autonomous agent loop suffer from state leakage during cold starts?▼

State leakage during cold starts occurs when autonomous agent loops lack external state files. Enforcing a blueprint that mandates separate state tracking prevents this failure, ensuring the agent resumes correctly without corrupting previous execution data.