skills.rest
rp4ri
rp4riCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

dev

Orchestrate multi-agent bug-hunting workflows across codebases with role-based execution.

Updated Mar 15, 2026
Search Tags:#security#multi-agent#code-review#scanning#bug-hunt
One-click install
npx skills add https://github.com/rp4ri/claude-local-marketplace --skill dev-rp4ri
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: dev
Source: https://github.com/rp4ri/claude-local-marketplace/tree/main/plugins/dev-studio/skills/dev
Command: npx skills add https://github.com/rp4ri/claude-local-marketplace --skill dev-rp4ri

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

Development teams often face slow, brittle bug-hunting cycles that miss subtle runtime issues. This Skill orchestrates a multi-agent pipeline to systematically find, challenge, and fix real bugs across large codebases, improving confidence and throughput.

Core Features & Use Cases

  • Orchestrates Recon, Hunter, Skeptic, Referee, and Fixer roles in a sequential-first workflow with optional parallel tasks for read-only steps.
  • Reuses local artifacts (.bug-hunter/*) for state, triage, threat modeling, and reports to maintain continuity across runs.
  • Enables safe, iterative bug fixing with canaries, circuit breakers, and rollbacks, suitable for PR reviews, security audits, and dependency scanning.

Quick Start

Run the end-to-end bug-hunting flow using local-sequential mode and review the final report in .bug-hunter/report.md.

Frequently Asked Questions about dev

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

FAQPage Schema
How do I automate multi-agent code review to find runtime bugs?▼

Automate multi-agent code review by orchestrating sequential roles—Recon, Hunter, Skeptic, Referee, and Fixer—to systematically discover, challenge, and fix runtime bugs across your codebase.

What is a multi-agent bug-hunting workflow for security scanning?▼

A multi-agent bug-hunting workflow coordinates specialized roles to read local triage data, challenge potential threats, and fix real runtime bugs safely across large codebases.

How do I run safe iterative bug fixing with rollbacks during dependency scanning?▼

Run safe iterative bug fixing using built-in canaries, circuit breakers, and rollbacks to manage dependency scanning and apply fixes without destabilizing the codebase.

Can I reuse triage and threat modeling state across multiple bug-hunting runs?▼

Yes, reuse triage and threat modeling state across runs by reading and writing local artifacts within the .bug-hunter directory to maintain full workflow continuity.

Does the bug-hunting workflow support parallel tasks for read-only code scanning?▼

Yes, the bug-hunting workflow uses local-sequential execution by default but supports optional subagents to run read-only code scanning and recon steps in parallel.

When should I not use a sequential-first multi-agent pipeline for bug hunting?▼

Avoid a sequential-first multi-agent pipeline when you need real-time interactive debugging or lack local prompt references and .bug-hunter artifacts to guide the execution flow.

Related Skills

Explore More Software Engineering Skills→
codexstar69codexstar69
479

threat-model-generation

Generate STRIDE-based threat models and security configurations for code repositories.

Community
Intermediate
codexstar69codexstar69
479

skeptic

Review reported vulnerabilities in code to identify false positives.

Community
Advanced
RangeyRoverRangeyRover
3

speckit.checker

Run static analysis across multiple languages and generate prioritized issue reports.

Community
Intermediate
galatanovidiugalatanovidiu
2

hyper-code-review

Automates a structured three-pass code review of diffs.

Community
Advanced