launch-native

Route planning, development, and QA tasks through native Agents.

6|Updated Apr 8, 2026
One-click install
npx skills add https://github.com/withpiper/pipekit --skill launch-native
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: launch-native
Source: https://github.com/withpiper/pipekit/tree/main/skills/launch-native
Command: npx skills add https://github.com/withpiper/pipekit --skill launch-native

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Route planning, development, and QA tasks through native Agents to evaluate the necessity of VBW as an intermediate layer.

Core Features & Use Cases

  • VBW-free planning flow: Use native Plan/Review/Execute/Verify subagents directly, bypassing vbw-lead, vbw-dev, and vbw-qa.
  • Spike comparison: Run /launch-native alongside /launch on the same Linear issue to compare plan quality, execution fidelity, and gating behavior.
  • Contextual context: Read project context from method.config.md and write plans to a simplified location, enabling quick evaluation without VBW scaffolds.

Quick Start

Run /launch-native on a Linear issue and compare its outputs against /launch, using the same issue context.

Frequently Asked Questions about launch-native

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

FAQPage Schema
How do I evaluate native agents without a VBW intermediate layer for planning and QA?

To evaluate native agents without a VBW intermediate layer, route planning, development, and QA tasks directly through native Plan, Review, Execute, and Verify subagents, bypassing vbw-lead, vbw-dev, and vbw-qa scaffolds entirely.

What is the best way to compare native agent execution fidelity against a standard launch on Linear issues?

The best way to compare execution fidelity is running a spike scenario where you execute /launch-native alongside /launch on the same Linear issue, measuring plan quality, execution fidelity, and gating behavior differences side by side.

Do I need specific project context to test native agents as a launch spike?

Yes, testing native agents requires the same project context provided via a method.config.md file. This contextual configuration ensures both /launch and /launch-native read identical project parameters for accurate comparison.

How does native agent planning store outputs for quick evaluation?

Native agent planning stores outputs in a simplified location designed for quick evaluation. This approach removes VBW scaffolds from the output path, enabling faster review of plan quality and gating behavior.

When should I use native agents instead of VBW subagents for development tasks?

Use native agents instead of VBW subagents when you need to run spike scenarios to measure the necessity of VBW as an intermediate layer. It helps evaluate baseline plan quality and execution fidelity without scaffolding.