test-fix-loop-N

Distribute failing Go tests across parallel agents for fixes and verification.

2|Updated Feb 24, 2026
One-click install
npx skills add https://github.com/lestrrat/claude-code --skill test-fix-loop-n
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: test-fix-loop-N
Source: https://github.com/lestrrat/claude-code/tree/main/skills/test-fix-loop-N
Command: npx skills add https://github.com/lestrrat/claude-code --skill test-fix-loop-n

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Discover Go test failures, partition by package/count, and spawn N parallel test-fix-loop agents to verify results against a shared parent branch.

Core Features & Use Cases

  • Discovery of failing tests across the repository and grouping by package for efficient targeting.
  • Parallel orchestration of agents to run fixes and validation without cross-contamination.
  • Unified verification against a shared parent branch using the existing sync protocol.

Quick Start

Start by choosing a parent branch and running the orchestrator to initiate a wave of agents.

Frequently Asked Questions about test-fix-loop-N

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

FAQPage Schema
How do I run parallel Go test fixes across multiple agents?

To run parallel Go test fixes, this Skill discovers failing tests, partitions them by package, and spawns up to N parallel agents using isolated worktrees to apply fixes and verify results against a shared parent branch.

How does test partitioning by package work for parallel debugging in Go?

Test partitioning by package groups discovered failing Go tests, distributing each package subset to a dedicated agent to ensure isolation and prevent cross-contamination during parallel debugging.

Do I need git worktrees to isolate parallel Go test runs?

Yes, git worktrees are required to establish isolated agent contexts. Each partitioned agent operates within its own worktree to apply fixes independently before consolidating reports for final verification.

What is the best way to consolidate Go test fixes from multiple agents?

The best way to consolidate Go test fixes is using a central orchestrator that collects per-agent reports and drives a unified verification step against the required parent branch.

Can I use this parallel testing approach without specifying a parent branch?

No, a parent branch argument is required. The orchestrator enforces a shared parent branch to verify all distributed agent fixes and ensure final consolidation aligns with the correct codebase state.