pelizzai-loop

Execute multi-round tasks through OODA loops until a verified Definition of Done.

Updated Jun 26, 2026
One-click install
npx skills add https://github.com/rpelizza/PelizzAI --skill pelizzai-loop
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: pelizzai-loop
Source: https://github.com/rpelizza/PelizzAI/tree/main/.agents/skills/pelizzai-loop
Command: npx skills add https://github.com/rpelizza/PelizzAI --skill pelizzai-loop

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill prevents AI agents from drifting, quitting prematurely, or iterating indefinitely by enforcing a structured, evidence-based loop until a verified Definition of Done is reached.

Core Features & Use Cases

  • OODA Loop Enforcement: Structures work into Observe, Orient, Decide, and Act cycles to ensure every step is grounded in current reality.
  • Verified Completion: Mandates that the Definition of Done is met and verified with fresh evidence before concluding any task.
  • Stop Criteria: Provides clear, legitimate exit points for material decisions or blockers, ensuring the human is consulted only when necessary.
  • Use Case: Use this when managing complex, multi-round investigations or multi-task development plans where you need to ensure the agent stays on track and validates every outcome.

Quick Start

Initiate the Pelizzai Loop to execute the current development plan until the Definition of Done is verified.

Frequently Asked Questions about pelizzai-loop

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

FAQPage Schema
How do I stop AI agents from drifting or quitting prematurely during complex task execution?

To prevent premature quitting, use OODA-based task execution loops that enforce evidence-based validation against a verified Definition of Done, ensuring agents iterate until stop criteria are legitimately met.

What is the OODA loop framework for AI workflow automation?

The OODA loop framework for AI workflow automation structures work into Observe, Orient, Decide, and Act cycles, ensuring every step in multi-round investigations or iterative bug-fix cycles is grounded in current reality.

How do I manage multi-task development plans with verified completion criteria?

Manage multi-task development plans by applying an OODA-based execution loop that requires fresh evidence to verify the Definition of Done, providing clear exit points for material decisions or blockers.

When should I use structured task execution loops for multi-round investigations?

Use structured task execution loops for multi-round investigations when you need to ensure consistent progress toward a verified Definition of Done and want to prevent indefinite iteration or premature abandonment.

Does OODA-based task management work for iterative bug-fix cycles?

OODA-based task management works for iterative bug-fix cycles by driving macro-level execution through Observe, Orient, Decide, and Act phases, ensuring evidence-based validation before the workflow concludes.

What are the limitations of using OODA loops for AI task management?

A limitation of OODA loops for AI task management is the strict adherence required to explicit stop criteria; the loop pauses for human consultation only when material decisions or legitimate blockers are encountered.