plugin-mh
01Plugin Claude Code
Claude Code plugins for thinking, deciding, and building smarter by MH.
Plugin Claude Code
Claude Code plugins for thinking, deciding, and building smarter by MH.
Plugin Claude Code
A collection of custom Claude Code skills and agents created by MH. The description does not explain what individual skills or agents do.
Instructions file CodexOpenCode
An AGENTS.md instruction file for the kingkingburger/plugin-mh project. It documents the project, how to build, test, and run it, its directory structure, protected areas, and coding conventions.
Instructions file
A set of Claude Code instructions for the plugin-mh project, including its structure, agents, skills, and removed skills.
Agent
Expert code review specialist with severity-rated feedback, logic defect detection, SOLID principle checks, style, performance, and quality strategy.
Agent
Knowledge curator that extracts reusable learnings, patterns, preferences, and follow-ups from a work session and accumulates them into auto-memory. Filters noise so only durable, non-obvious value persists. Optionally emits a markdown report.
Skill Claude CodeCodex
This skill should be used when the user wants multiple AI agents to debate a topic from different perspectives. Trigger on "agent-arena", "에이전트 토론", "에이전트 싸워", "debate this", "여러 관점에서 분석", "찬반 토론", "agent debate", "arena", "팀 토론", "관점별 분석". Spawns role-assigned agents that argue in multiple rounds, then synthesizes…
Agent
An agent that argues strongly in favour of a topic, focusing on its benefits and opportunities.
Agent
An assessment agent that reviews implementation feasibility, technical debt, and scalability. It is included in the 5/8 preset.
Agent
A review agent that examines weaknesses, risks, and ways a project could fail. The description is in Korean and gives no further details about its specific methods.
Agent
A planning agent that judges schedules and resources against expected return on investment (ROI), meaning the value gained compared with the cost.
Agent
An edge-case testing agent that looks for scenarios likely to break or behave unexpectedly.
Agent
An agent that combines opinions from other agents and produces options with a recommendation.
Agent
A user-experience review agent that evaluates work from the user's point of view.
Agent
An agent that provides long-term vision and strategic direction. The description is in Korean and says it is included only in an eight-agent preset.
Skill Claude CodeCodex
A Korean-language cleanup workflow for removing low-quality or unnecessary AI-generated code, with tests used to guard against regressions, meaning new changes that break existing behavior.
Skill Claude CodeCodex
An automatic commit workflow that checks the changes made by a task and groups related changes by type before creating English Conventional Commit messages. Conventional Commits are a standard format for describing changes in Git.
Skill Claude CodeCodex
A lightweight session-closing skill that saves important issues, lessons, and preferences to the agent’s persistent memory.
Skill Claude CodeCodex
An audit of accumulated skills, agents, instruction files, prompts, memory, and project documents. It checks for conflicts, excessive searching, and drift between source and distributed copies.
Skill Claude CodeCodex
Relentlessly interview the user about a plan or design until the decision tree is clear. --docs mode additionally grounds every question in the repo's domain model, glossary, and ADRs, then updates CONTEXT.md. Trigger on "grill me", "grill-me", "grillme", "계획을 까다롭게 질문해줘", "설계 압박 질문", "stress-test my plan", "grill with…
Skill Claude CodeCodex
A workflow for checking that implemented features work by using them like a real user. It includes planning and running end-to-end tests, which test a feature across the whole application.
Skill Claude CodeCodex
A staged process for turning unclear ideas into detailed requirements, a design, checks, and then implemented stories. It uses checklists and review steps before moving forward.
Skill Claude CodeCodex
A staged code-review loop that starts with one reviewer and brings in an architect and critic only when medium-or-more serious findings appear.
Skill Claude CodeCodex
Skill "skill-manage" from kingkingburger/plugin-mh, covering skill-manage: plugin-mh skill manifest sync, 관리 대상 파일 (manifest files), when to use, mode 1: delete and step 1 — 대상 확인.