Red-test bug hunt
Quest Reproduce a bug, fix it and keep a test that guards it
- Hero
- Codex
- Weapon
GPT-6 Sol- Charge
Sentry MCP
Superpowers Systematic Debugging
Matt Pocock TDD
Microsoft Playwright MCP
Hunter skill tree
Every spell in this class, in the row where it usually goes. Pick a few from each row for a loadout.
ScoutGather context
Nothing equipped here
CraftMake the thing
Nothing equipped here
TestCheck the work
- Superpowers Systematic DebuggingScroll
Superpowers Test-Driven DevelopmentScroll
Matt Pocock Diagnosing BugsScroll- Matt Pocock TDDScroll
Anthropic Webapp TestingScroll
Addy Osmani Browser Testing with DevToolsScroll
Addy Osmani Debugging and Error RecoveryScroll
Chrome DevTools MCPOrb
Unit Testing (wshobson agents)Relic- Microsoft Playwright MCPOrb
Trail of Bits Mutation TestingRelic
OpenAI Frontend Testing and DebuggingScroll- Sentry MCPOrb
Hegel property-based testingRelic
Trunk Flaky TestsRelic
ShipDeliver it
Nothing equipped here
Loadouts
Combos
Pin the bug, watch it live
Hunter · Testing and debugging- Matt Pocock TDD
- Anthropic Webapp Testing
- Chrome DevTools MCP
Reproduce a reported error
Hunter · Testing and debugging- Sentry MCP
- Superpowers Systematic Debugging
- Microsoft Playwright MCP
Test-first bug hunt
Hunter · Testing and debugging- Superpowers Systematic Debugging
- Matt Pocock TDD
Spells
15 spells
Finds a root cause and a failing test before any fix, and stops after three failed attempts.
302,865 GitHub stars on obra/superpowers
Requires a failing test before implementation code for any feature or bugfix.
302,865 GitHub stars on obra/superpowers
Builds a tight pass/fail loop for a hard bug or slowdown before changing the suspected code.
299,281 GitHub stars on mattpocock/skills
Test-first development through public interfaces, agreed seams, and one behavioral slice at a time.
299,281 GitHub stars on mattpocock/skills
Lv 6
Writes Python Playwright scripts to test local web apps, with a server helper.
182,629 GitHub stars on anthropics/skills
Inspects a live page through Chrome DevTools MCP: DOM, console, network, and performance.
109,146 GitHub stars on addyosmani/agent-skills
Stops feature work on a failure, keeps the evidence, and traces the root cause before editing.
109,146 GitHub stars on addyosmani/agent-skills
Drives live Chrome for clicks, console, network, screenshots, and performance traces.
53,926 GitHub stars on ChromeDevTools/chrome-devtools-mcp1,737,496 npm downloads last week for chrome-devtools-mcp
Lv 5
Claude Code plugin with a test-generate command plus test automator and debugger subagents for Python and JavaScript unit tests.
41,508 GitHub stars on wshobson/agents
Browser automation and inspection through Playwright and structured accessibility snapshots.
38,460 GitHub stars on microsoft/playwright-mcp3,465,430 npm downloads last week for @playwright/mcp
Skill that sets up mewt or muton mutation campaigns, sorts surviving mutants into test gaps or equivalents, and checks bugs they point to.
7,705 GitHub stars on trailofbits/skills
Checks a rendered page with the Browser plugin, or Playwright if Browser is missing.
7,519 GitHub stars on openai/plugins
Pulls Sentry issues, errors, traces, and project context into the agent over OAuth.
936 GitHub stars on getsentry/toolkit
Teaches the agent to write and review property-based tests with Hegel in Rust, Go, C++, TypeScript, Java and OCaml projects.
85 GitHub stars on hegeldev/hegel-skill
Claude Code plugin with a flaky test patterns skill, plus Trunk commands that explain and fix flaky tests from your CI history.
0 GitHub stars on trunk-io/claude-code-plugin