factory-preventive-sweep

Scan codebases for runtime defect patterns and output severity-ordered reports.

1|Updated Apr 8, 2026
One-click install
npx skills add https://github.com/e2its/myrmion-AI-factory --skill factory-preventive-sweep
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: factory-preventive-sweep
Source: https://github.com/e2its/myrmion-AI-factory/tree/main/.claude/skills/factory-preventive-sweep
Command: npx skills add https://github.com/e2its/myrmion-AI-factory --skill factory-preventive-sweep

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Static build verification loops (linting, type checking, static application security testing, and unit tests) cannot detect defects that only manifest under real runtime conditions with live authentication providers, databases, content delivery networks, and browsers, leading to costly post-deployment failures and user impact.

Core Features & Use Cases

  • Parallel Scope-Based Scanning: Spawns one sub-agent per non-overlapping system scope (backend, frontend, infrastructure, cross-cutting) derived dynamically from the project's Defect Prevention Catalog, with no hardcoded concurrency limits.
  • Feature-Scope Filtering: Automatically excludes irrelevant defect checks based on the feature's defined scope (frontend-only, backend-only, integration, full-stack) to reduce unnecessary scan overhead.
  • Integrated Governance Workflow: Consolidates findings into a severity-ordered report, requires user approval before fixes are applied, and saves a structured report artifact to the feature's review directory for auditability.

Quick Start

Invoke the factory-preventive-sweep skill to scan your recently deployed feature for runtime defects that were not caught by static build gates, and review the consolidated severity-ordered report before applying fixes.

Frequently Asked Questions about factory-preventive-sweep

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

FAQPage Schema
How do I catch runtime defects that static build gates miss after deployment?

To catch runtime defects missed by static build gates, you can run a preventive sweep that searches for empirically discovered defect patterns across backend, frontend, and infrastructure codebases after deployment.

What is a preventive defect sweep and when should I run it?

A preventive defect sweep scans full-stack features for runtime issues invisible to linting or unit tests. Run it on first feature deployments, after core architectural changes, or during post-major fix batch scans.

How do I scan full-stack features for runtime defects without hardcoded concurrency limits?

You can scan full-stack features by spawning parallel sub-agent searches per system scope. Scopes are dynamically derived from the project's Defect Prevention Catalog to ensure comprehensive coverage without concurrency limits.

Does this preventive validation work for frontend-only features or is it limited to full-stack?

Preventive validation works for frontend-only, backend-only, integration, and full-stack features. It automatically excludes irrelevant defect checks based on the feature's defined scope to reduce scan overhead.

Why do static application security testing and unit tests fail to prevent post-deploy failures?

Static application security testing and unit tests fail post-deploy because they cannot detect defects that manifest under real runtime conditions with live databases, authentication providers, content delivery networks, and browsers.

How are runtime defect scan results consolidated for remediation?

Runtime defect scan results are consolidated into a severity-ordered report requiring user approval before fixes are applied. A structured report artifact is saved to the feature's review directory for auditability.