noob-visual-pool

Coordinate visual QA pool agents to run tests without race conditions.

Updated Apr 4, 2026
One-click install
npx skills add https://github.com/ganeshgaxy/apps-for-testing --skill noob-visual-pool
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: noob-visual-pool
Source: https://github.com/ganeshgaxy/apps-for-testing/tree/main/.claude/skills/noob-visual-pool
Command: npx skills add https://github.com/ganeshgaxy/apps-for-testing --skill noob-visual-pool

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Run visual QA pool agents for a ticket. Checks qa_pool_agents for config, creates a visual run, populates entries, enumerates pending visual test cases (filtering already-claimed), then launches one sub-agent per test case (round-robin across agent configs). Each sub-agent claims its assigned test case by name via --name flag — no race conditions.

Core Features & Use Cases

  • Checks qa_pool_agents for configuration, creates and populates a visual run, enumerates pending entries (excluding already-claimed), then distributes test cases across agent configs using a round-robin approach.
  • Pre-claims each pending entry to ensure deterministic assignment and avoid race conditions.
  • Launches a separate sub-agent per test case, with mode-aware runs (baseline or verification) and per-agent customization via agent configs.

Quick Start

Use this skill to coordinate a visual QA pool for a ticket and spawn agents to execute pending visual test cases.

Frequently Asked Questions about noob-visual-pool

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

FAQPage Schema
How do I run visual tests in parallel without race conditions?

Visual test parallel race conditions are prevented by atomically pre-claiming pending entries before launching any sub-agents. This deterministic assignment ensures multiple test cases evaluate concurrently across various agent configurations without conflicts over the same ticket.

How do I distribute visual QA test cases across multiple agent configs?

Visual QA test cases are distributed across multiple agent configs using a round-robin assignment approach. The system enumerates pending entries, filters already-claimed tests, and launches one sub-agent per test case to ensure balanced parallel execution.

What is atomic pre-claiming in a visual QA pool workflow?

Atomic pre-claiming in a visual QA pool is a mechanism to deterministically lock pending test entries before execution. It prevents race conditions by ensuring each sub-agent claims its assigned test case by name via a flag, avoiding overlapping evaluations.

Can I configure a spawn cap for parallel visual test agents?

Yes, you can configure a spawn cap for parallel visual test agents. The coordination system supports a configurable spawn cap to limit the number of active sub-agents, alongside mode-aware visual runs and dashboard-based monitoring for the QA pool.

How do I monitor pending entries during a visual test run?

You monitor pending entries during a visual test run using dashboard-based monitoring. The system creates a visual run, populates entries, and tracks the status of pending and claimed test cases across the distributed agent configurations.