sus

Detect suspicious architectural and maintainability issues across codebases with prioritized reports.

5|Updated Feb 17, 2026
One-click install
npx skills add https://github.com/Nathan13888/nice-skills --skill sus
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sus
Source: https://github.com/Nathan13888/nice-skills/tree/main/skills/sus
Command: npx skills add https://github.com/Nathan13888/nice-skills --skill sus

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Identifies suspicious, architecturally problematic, and high-impact maintainability issues across a codebase by orchestrating focused analysis agents and synthesizing their output into a prioritized actionable report.

Core Features & Use Cases

  • Parallel agent analysis: Deploys modular agents (architecture, complexity, coupling, consistency, risk) to scan the repository in parallel and surface targeted findings.
  • Compilation gate and safety checks: Verifies the project builds before analysis and requires explicit user input when no build system is detected.
  • Consolidated, prioritized reporting: Deduplicates cross-agent findings, ranks by severity and agreement, and produces file-level locations with remediation guidance.
  • Use Case: Run against a product repository before a major release to catch architectural drift, high-risk error handling, and maintainability hotspots that warrant immediate attention.

Quick Start

Run the sus skill to audit the repository and produce a prioritized report of suspicious architectural and maintainability issues with file-level locations and suggested fixes.

Frequently Asked Questions about sus

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

FAQPage Schema
How do I find architecturally problematic code issues in my repository?

Static analysis for technical debt identifies suspicious, high-impact maintainability issues by scanning repository structure, complexity, and coupling. It deduplicates cross-agent findings and ranks them by severity to produce a prioritized, actionable report.

Can I run static analysis on a codebase if no build system is detected?

Running static analysis without a detected build system requires explicit user input to proceed. The analysis uses a compilation gate to verify the project builds before deploying configurable agents to inspect your codebase.

What is the best way to assess technical debt and risk before a major release?

Assessing technical debt before a major release is best handled by deploying modular analysis agents to catch architectural drift and high-risk error handling. This approach surfaces maintainability hotspots that warrant immediate attention with suggested fixes.

How does parallel agent analysis work for code review and risk assessment?

Parallel agent analysis for code review works by deploying modular agents to scan the repository simultaneously, targeting architecture, complexity, coupling, consistency, and risk. It deduplicates cross-agent findings to produce prioritized reports.

Does automated code review provide specific file locations for maintainability issues?

Automated code review does provide specific file-level locations for maintainability issues. It deduplicates findings across analysis agents, ranks them by severity and agreement, and includes actionable remediation guidance.