bug-discovery-pro

Identify and triage evidence-backed defect candidates in large codebases.

1|Updated Jul 3, 2026
One-click install
npx skills add https://github.com/truongnat/skills --skill bug-discovery-pro
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: bug-discovery-pro
Source: https://github.com/truongnat/skills/tree/main/skills/bug-discovery-pro
Command: npx skills add https://github.com/truongnat/skills --skill bug-discovery-pro

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Triages and surfaces candidate defects in large codebases using a graph-based workflow to reveal hidden bugs and their impact.

Core Features & Use Cases

  • Graph-driven triage: Use GitNexus to map symptoms to code paths and identify blast radius.
  • Evidence-backed candidates: Generate a list of defect candidates with confidence levels and references to evidence.
  • Workflow integration: Coordinate with testing-pro and security-pro for handoffs; supports re-indexing and edge cases.
  • Guided remediation: Provide structured outputs to assist triage and planning.

Quick Start

Run a repository-wide scan to surface bug candidates and review the evidence with GitNexus before proceeding to remediation.

Frequently Asked Questions about bug-discovery-pro

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

FAQPage Schema
How do I triage bug candidates in a large codebase?▼

Triage bug candidates in a large codebase by using a graph-based workflow to map symptoms to code paths. This approach surfaces evidence-backed defects and assesses the blast radius for related bugs.

What is graph-driven code analysis for finding hidden defects?▼

Graph-driven code analysis uses GitNexus to map symptoms to code paths and identify blast radius. It generates a structured list of defect candidates complete with confidence levels and references to evidence.

How do I assess the blast radius of related bugs across my codebase?▼

Assess the blast radius of related bugs by applying MCP tools in a defined order to map symptoms to code paths. This reveals hidden bugs and their impact across the repository.

Can I use this graph workflow to scan my entire repository for defects?▼

Yes, you can run a repository-wide scan to surface bug candidates. The workflow applies structured MCP tools to map symptoms and produce evidence-backed defect candidates for remediation.

What is the best way to map code symptoms to specific code paths?▼

The best way to map code symptoms to code paths is applying MCP tools in a defined order including query, context, impact, and shape_check. This generates evidence-backed candidates with recommended next steps.

How do I handoff triaged bug candidates to security and testing workflows?▼

Handoff triaged bug candidates by generating structured outputs with evidence and confidence levels. The workflow explicitly coordinates handoffs to testing-pro and security-pro when appropriate.