Plugins

from OpenSPP/claude-plugins

Spec-driven offensive-security framework: 31 kill-chain skills, executable safety controls, pattern-learning memory, bounded engagement engine, and a SessionStart skill-invocation dispatcher.

Create and edit Office documents — docx, pdf, pptx, and xlsx

from FradSer/dotclaude

Office of Claudette, Adriaan's personal assistant. The task skill is the single human interface to the canonical local task store: add, edit, complete, cancel and maintain tasks against the store contract, with full history and logging. Writes only the store and the logs, never Outlook, and never deletes.

Bundles the offshore MCP server. Forwards refused prompts to a local abliterated model (Ollama, LM Studio, vLLM, or any OpenAI-compatible endpoint) and returns the unrestricted reply.

Portable orchestrator: generic agents, skills, guard hooks, and /og:orchestrate + /og:orchestrate-init slash commands for any repo.

大喜利（Ogiri）の回答を生成するスキル

Ohio court documents: Ohio Civ. R. 10 + per-court local rules (verbatim Cuyahoga + Delaware Common Pleas); eight Court of Common Pleas venues + municipal-courts + family-court; five subject bundles — consumer-debt, family-law, personal-injury, employment, commercial-disputes. Full detail in the plugin README.

Plays the FAAAH sound when your Claude Code context window usage hits 80%. Never get surprised by context limits again.

Java Spring Boot 멀티모듈 프로젝트를 위한 AI 기반 개발 플러그인

Personality-driven buddy reactions, XP progression, achievements, commit garden, and chaos-fueled rage meters. 18 species, 5 stats, every buddy feels different.

from 2lab-ai/oh-my-claude

from zephyrpersonal/oh-my-claude-code

OMO-slim orchestration port for Claude Code: SessionStart directive + per-turn anchor hooks, five specialist subagents (explorer, librarian, oracle, designer, fixer), and ported OMO skills.

Claude Code native multi-agent orchestration with intelligent model routing, 28 agent variants, and 32 powerful skills. Zero learning curve. Maximum power.

GX 사업본부 개발자를 위한 개발 자동화 플러그인. PRD, 설계, 구현, 리뷰, PR까지 에이전트 팀이 처리합니다.

Zero-config harness: spec-driven autonomous loop (self/cross-verification, tiered guardrails) + weight-aware routing & multi-model independent verify, convention detection, test enforcement, model routing, ambiguity guard, dangerous guard, multi-agent with git worktrees, native team orchestration

Give your AI coding agent a permanent second brain. Connect Claude to an Obsidian vault to auto-record decisions and recall context.

A compact Claude Code statusline focused on context window usage and compact timing.

Implementation planning: read a spec, propose architecture, structured self-review, deliver a plan document

Requirement thinking: brainstorm, clarify ambiguities, define acceptance criteria, deliver a spec document

디자인 자동화 스킬 모음: htmlviaTrueSource(HTML→스펙·프로토타입·QA), new-ods-ui-builder(Figma→ODS HTML + pixel-diff), figma-to-ods-react(Figma→ODS React 화면 변환 + MCP 하이브리드)

Plugin from openjunto/oj-claude

Suite upkeep dispatcher — discovers which ok-plugins a project integrates and drives each one's own doctor/affirm verbs; knows the integration contract, never plugin internals

The ok-planner documentation corpus — concepts, provable stories and decisions, an issue queue — plus the sprint planning ceremony and corpus verbs (audit, prove)

Code standards for agentic maintenance — ships the Plumbline methodology: rules, lint binary, PostToolUse hook, and compliance skills

Workspace hygiene for parallel agent work — worktree-per-job, isolated runtime stacks, content-addressed artifacts — tailored to the project's declared stack profile

Open Knowledge Format (OKF) tooling. Author, initialize, validate, and refactor knowledge into vendor-neutral OKF v0.1 bundles: markdown files with YAML frontmatter.

OKF knowledge-bundle skill and lifecycle commands.