Prompts
Prompt engineering and templates
anytype
by CodeDraig
Enables Claude to create and manage objects in Anytype via Playwright MCP
parallelize
by biruk741
Decompose a converged plan into parallel execution phases. Generates PLAN.md and ORCHESTRATOR.md files with wave assignments, complexity tiers, and dependency graphs. Must be run after /plan when alignment is reached.
avp
by portofcontext
Use this skill for ANY work on AVP (Agent Voyager Project): agents, supervisor Commissions, conformance cases, or debugging AVP runs. AVP wires agents (drive an LLM, emit events) to supervisors (declare environment).
newsletter-assembly
by arielperez82
Assemble selected stories, poll, and show notes into a complete newsletter
skill-structure-audit
by rj-yingjunjie
审核并规整 Claude Code 技能(Skill)或插件(Plugin)的项目结构，引导用户从零创建符合规范的技能/插件目录
preqstation
by sonim1
Fetch tasks from PREQSTATION, execute work, and push updates/results back to PREQSTATION APIs. Use when asked to run a PREQ task, fetch todo tasks, or update execution status. Requires PREQSTATION_API_URL and PREQSTATION_TOKEN environment variables. Tasks carry an engine field (claude codex gemini) indicating which AI agent should execute them.
SKILL.md – Agentic Medical Device Reviewer（FDA / TFDA 審查官輔助）
by Erin100925
模型適合作為「整理與提示」，真正的批准/不批准、補件要求仍需由審查官判斷。
LintGate
by rohanvinaik
For Claude Code, all three files work together. For other agents, AGENTS.md is the primary entry point and integrate.sh generates an appropriate cognitive context file.
changeset-validation
by aiagentskills
Validate changesets in openai-agents-js using LLM judgment against git diffs (including uncommitted local changes). Use when packages/ or .changeset/ are modified, or when verifying PR changeset compliance and bump level.
oracle
by charys117
Prepare an "ask an expert" handoff bundle: a pasteable prompt.md plus a context.zip containing the minimum repo files needed for a grounded second opinion from an external assistant (e.g., ChatGPT Pro / GPT-5.2 Pro). Use only when the user explicitly asks to "use Oracle", "ask ChatGPT Pro", "ask an expert", or "get a second opinion". Never include secrets (.env, keys, credentials) in the bundle.
ai-wrapper-product
by ncdevshiv
"Expert in building products that wrap AI APIs (OpenAI, Anthropic, etc.) into focused tools people will pay for. Not just 'ChatGPT but different' - products that solve specific problems with AI. Cov..."
create-worker
by Cygnusfear
This skill should be used when creating custom agents for Claude Code, configuring specialized AI assistants, or when the user asks about agent creation, agent configuration, or delegating tasks to workers. Covers both file-based agents and teams delegation.
setup-claude
by AnsarUllahAnasZ360
Interactive Claude Code repository setup and optimization. Configures the complete ecosystem - skills, commands, subagents, hooks, rules, MCPs, and plugins. Invoke with /setup-claude init or /setup-claude audit.
Skill Security Scanner
by RekitRex21
Recommendation (Safe / Review / Block)
reclaude
by artivilla
Refactor CLAUDE.md files to follow progressive disclosure principles. Use when CLAUDE.md is too long or disorganized.
ReadMate Pipeline — SKILL.md
by PU3-Lab
OCR: PaddleOCR 단독 → PaddleOCR + Surya 혼합 (복잡 레이아웃)
vibenotifications
by pooriaarab
"Customizable notifications for Claude Code -- GitHub PRs, Slack DMs, stock prices, and more while you code"
iron-canvas-master
by Navigata1
Iron Canvas is an adaptive, site-aware, multi-agent design enhancement and creation methodology. It is not a template — it is an orchestration system. Three lineages fused into one definitive methodology: (A) Island Development Crew production pipeline — 7-phase battle-tested system from real projects: Oakwood AI Hub, Crown Collection, Iron Canvas Studio site. Failure taught Phase 1 (uniform templates killed site identity). Success encoded everything else. (B) Rigorous animation architecture — Brand Personality Matrix, GSAP motion taxonomy, Awwwards 5-axis verification, AI Reference Scout with North Star image generation. (C) Antigravity multi-agent orchestration pattern — parallel Build Agents, Cross-LLM Technical Validation, Design PRD Packaging, Expertise Injection Protocol, and Router-Worker swarm architecture for maximum throughput. Three modes: SOLO: One agent, full pipeline, sequential execution. Slower, simpler to orchestrate. SWARM: Orchestrator + up to 5 parallel Build Agents. Faster, higher quality. MISSION: Full team — Orchestrator + Research + Validation + 5 Build Agents + QA. Fastest. Trigger for ANY digital design task: websites, dashboards, apps, scroll animations, mission controls, landing pages, or brand experiences. Runs natively in: OpenClaw, Antigravity Agent Manager, Claude Code, any agentic AI.
crafting-instructions
by arielperez82
Generate optimized instructions for Claude (Project instructions, Skills,
clack
by ahmadawais
Comprehensive guidance for building and debugging JavaScript and TypeScript CLIs with @clack/core and @clack/prompts, backed by bundled upstream source and examples. Use when implementing clack prompt flows, deciding between core primitives and styled prompts, adapting clack examples, or validating clack API usage.
githooks
by rozwer
Claude Code 固有の git ガードを対話的ウィザードで設定します。
gemini
by junoh-moon
Interact with Google's Gemini model via CLI. Use when needing a second opinion from another LLM, cross-validation, or leveraging Gemini's Google Search grounding. Supports multi-turn conversations with session management.
user-confirm-skill
by maxoreric
v1.0.0 (2026-01-19): åå§çæ¬
md-to-docx-template
by jonnyschneider
Convert markdown to beautifully styled Word documents using custom templates. Supports branded fonts, colors, and table styling. Extract styles from existing docs or generate fresh templates.