skills.rest
gil0mendes
gil0mendesCOMMUNITY
Explore all skills by this author →
Website𝕏@gil0mendesGitHub

feedback-loop

Create deterministic feedback loops with reproduction, measurement, and exit criteria.

Updated Nov 3, 2019
Search Tags:#automation#testing#debugging#validation#data pipeline#feedback loop#ui/visual
One-click install
npx skills add https://github.com/gil0mendes/dotfiles --skill feedback-loop-gil0mendes
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: feedback-loop
Source: https://github.com/gil0mendes/dotfiles/tree/main/configs/opencode/skill/feedback-loop
Command: npx skills add https://github.com/gil0mendes/dotfiles --skill feedback-loop-gil0mendes

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill enables agents to create fast, deterministic feedback loops for self-validation, reducing stalled tasks and clarifying outcomes.

Core Features & Use Cases

  • Automated Validation: Integrates with tests, UI snapshots, and data metrics for continuous self-checking.
  • Reproduction & Measurement: Focuses on discovering or creating reproducible steps and measurable outcomes.
  • Use Case: When an agent is stuck on a bug, it can use this Skill to automatically run failing tests, hypothesize the cause, make a small code change, and re-run the test to confirm the fix, iterating until the outcome is clear.

Quick Start

Use the feedback-loop skill to debug a failing test by running the provided reproduction command and analyzing the output.

Frequently Asked Questions about feedback-loop

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

FAQPage Schema
How do I automate debugging when an agent gets stuck on a failing test?▼

Automated debugging uses a deterministic feedback loop to run failing tests, hypothesize the cause, apply a code change, and re-run the test to confirm the fix. This cycle iterates until the outcome is clear.

What is a deterministic feedback loop for self-validation?▼

A deterministic feedback loop is a self-validation process incorporating reproduction, measurement, and exit criteria. It enables agents to continuously self-check work against tests or metrics to clarify outcomes.

Can I use automated validation for UI visual tasks and data pipelines?▼

Yes, automated validation applies to UI visual and data pipeline scenarios. The feedback loop provides specific recipes for these tasks, integrating with UI snapshots and data metrics to measure outcomes.

How do I create reproducible steps for measuring agent outcomes?▼

You create reproducible steps by focusing on reproduction and measurement within the feedback loop. The agent discovers or creates reproducible commands, runs them, and analyzes the output to establish measurable outcomes.

Does the feedback loop approach work without external dependencies?▼

Yes, the feedback loop operates without external dependencies. It relies on an internal core loop of Sense, Hypothesize, Change, Measure, and Decide to facilitate self-validation across code, UI, and data pipeline tasks.

Related Skills

Explore More Software Engineering Skills→
hakushunhakushun

vitest

Run Jest-compatible unit tests for Vite projects with coverage and snapshots.

Community
Intermediate
getsentrygetsentry
44.5k

migrate-frontend-forms

Migrate frontend forms from JsonForm/FormModel to TanStack with code examples.

Official
Intermediate
sustinbebustinsustinbebustin

feedback-loop

Creates deterministic feedback loops for self-validation across code, UI, and data pipelines.

Community
Advanced
thoreinsteinthoreinstein
5

postmortem

Generates blameless postmortems with root-cause analysis for software incidents.

Community
Advanced