wv-detect-loop

Detects repeated failures and interrupts retry loops in AI-driven workflows.

Updated Feb 16, 2026
One-click install
npx skills add https://github.com/AGM1968/weave --skill wv-detect-loop
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: wv-detect-loop
Source: https://github.com/AGM1968/weave/tree/main/.claude/skills/wv-detect-loop
Command: npx skills add https://github.com/AGM1968/weave --skill wv-detect-loop

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Detects when repeated failures cause stagnation and automatically interrupts endless retries by steering the workflow toward alternative strategies.

Core Features & Use Cases

  • Detects loops in task attempts and time spent > estimates
  • Documents failed approaches and proposes 2-3 alternative strategies
  • Guides quick experiments with smallest falsifiable steps to regain progress
  • Useful for AI agents coordinating tasks across components, CI, and testing loops

Quick Start

Invoke the loop-breaker when you detect 2–3 consecutive failed attempts and reassess your approach.

Frequently Asked Questions about wv-detect-loop

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

FAQPage Schema
How do I break persistent retry loops in AI workflows?

Break persistent retry loops by detecting 2–3 consecutive failed attempts and automatically switching to alternative strategies. The Skill documents failed approaches and guides a strategic pivot to regain progress.

What is a strategic pivot in failure analysis?

A strategic pivot in failure analysis interrupts endless retries by documenting failed approaches, generating 2-3 alternative strategies, and selecting the smallest falsifiable experiment to test next.

When should I interrupt task attempts to reassess my approach?

Interrupt task attempts to reassess your approach when you detect 2–3 consecutive failed attempts, when edits recur, or when time spent exceeds initial estimates and progress stalls.

How do I generate alternative strategies after repeated workflow failures?

Generate alternative strategies after repeated workflow failures by documenting the failed approaches, then proposing 2-3 new pathways and selecting the smallest falsifiable experiment to quickly test viability.

Does loop detection work for CI and testing loops?

Loop detection works for CI and testing loops by identifying when progress stalls, edits recur, or time spent exceeds estimates across coordinated components, guiding a strategic pivot to break the cycle.