提示词
提示词工程与模板
write-cn
swaylq
写中文的时候直接绕开 AI 腔，而不是写完再洗。给 LLM 当写作指令用：喂进题目和文体， 产出的初稿就已经没有模板句式、没有价值拔高骨架、没有等厚段落。 这是六段流水线的第一段。第一段决定文章有没有人味，后面几段只能修，补不出来。 这份文件本身刻意写成它要求的那种中文 —— 指令文档的版式和语气会被模型当范文模仿， 一份满是加粗和编号清单的规范，教出来的就是加粗和编号清单。
cron
tkersey
Create and manage Codex app automations stored in the local SQLite database (~/.codex/sqlite/codex-dev.db). Use when you need to add, list, update, enable/disable, delete, or run-now automations; edit automation names, prompts, RRULE schedules, or cwd scopes; or inspect automation records while troubleshooting app automation behavior.
puff
tkersey
Launch and manage Codex Cloud tasks from the CLI, including detached background watchers that track completion. Use when users ask to run coding work in cloud/background agents, queue multiple cloud tasks, poll task status, fetch cloud diffs, apply cloud outputs locally, or pair cloud kickoff with $cas orchestration.
smart-illustrator
axtonliu
智能配图与 PPT 信息图生成器。支持三种模式：(1) 文章配图模式 - 分析文章内容，生成插图；(2) PPT/Slides 模式 - 生成批量信息图；(3) Cover 模式 - 生成封面图。所有模式默认生成图片，--prompt-only 只输出 prompt。支持 Bento Grid 功能展示图风格（--style bento）。触发词：配图、插图、PPT、slides、封面图、thumbnail、cover、bento grid、功能展示图、feature showcase。
Orquestador Data/ML
scaleto
Líder del Grupo Data/ML. Planifica y delega tareas de datos, machine learning, IA, LLMs y RAG a los especialistas del grupo.
orchestrating-swarms
Jamie-BitFlight
Facade skill for multi-agent swarm orchestration in Claude Code — routes to specialist skills covering primitives, spawning, operations, and patterns. Use when coordinating multiple agents in parallel, building pipeline workflows with dependencies, running parallel code reviews, creating self-organizing task queues, designing divide-and-conquer workloads, or choosing between TeamCreate and Agent tool approaches. Loads swarm-primitives (team lifecycle and message flow), swarm-spawning (agent types and backends), swarm-operations (tool API and error handling), and swarm-patterns (six orchestration recipes with complete workflows).
swarm-operations
Jamie-BitFlight
API reference for Claude Code multi-agent swarm tools -- TeamCreate, SendMessage, TeamDelete, TaskCreate/Update/List/Get, and Agent tool parameters. Use when looking up tool signatures, message schemas, shutdown sequences, error handling patterns, or debugging swarm operations. Covers direct messages, broadcasts, plan approval flows, graceful shutdown sequences, crashed teammate recovery, and common error causes.
skill-research-process
Jamie-BitFlight
Builds comprehensive Claude Code skills using parallel research agents — categorization, parallel documentation gathering, anti-hallucination checkpoints, and final validation. Use when building a skill from official docs, when "research for skill" or "create comprehensive skill" is requested, or when extensive multi-source documentation gathering is needed before skill creation.
cove-prompt-design
Jamie-BitFlight
Explains Chain of Verification (CoVe) prompt design — a 4-step pattern separating generation from independent factual verification. Use when designing prompts that require factual accuracy, reducing hallucinations, checking technical standards or APIs, or producing step-by-step procedures where subtle errors are costly.
code-review-claude-skills
Jamie-BitFlight
Loaded automatically when reviewing Claude skills or agent definitions — covers SKILL.md structure, frontmatter validity, token budget, description quality, and agent contract compliance.
code-review-llm
Jamie-BitFlight
Use when reviewing AI/ML code or LLM integration — activates on prompt templates, model selection logic, token budget concerns, or evaluation harness code. Enforces prompt hygiene, model tier matching, context window management, token economics, structured output validation, temperature settings, retry logic, streaming error handling, and PII/safety rules.
angular-directives
analogjs
Create custom directives in Angular v20+ for DOM manipulation and behavior extension. Use for attribute directives that modify element behavior/appearance, structural directives for portals/overlays, and host directives for composition. Triggers on creating reusable DOM behaviors, extending element functionality, or composing behaviors across components. Note - use native @if/@for/@switch for control flow, not custom structural directives.
skill-writer
metalagman
Use this skill to design, document, and structure AI agent skills for Gemini, Claude, and Codex. It provides architectural rules, directory standards, and writing best practices.
continuous-learning-v2
xu-xiang
基于直觉（Instinct）的学习系统，通过钩子（Hooks）观测会话，创建带有置信度评分（Confidence Scoring）的原子直觉，并将其演进为技能（Skills）、命令（Commands）或智能体（Agents）。
strategic-compact
xu-xiang
建议在逻辑间隔进行手动上下文压缩（Context Compaction），以便在任务阶段中保留上下文，而不是依赖随机的自动压缩。
continuous-learning
xu-xiang
自动从 Claude Code 会话（Sessions）中提取可重用的模式，并将其保存为学习到的技能以供未来使用。
design-trends-2026
guia-matthieu
"Apply 2026's top graphic design trends to any creative brief. Based on Kittl × Savee's 2026 Design Trends Report (10 trends + 2 honorable mentions), backed by Adobe, Figma, and Pinterest data. Use when: Designing a brand identity — pick the right aesthetic for your audience; Creating social media assets — use trending visual languages that perform; Briefing a designer or AI image tool — give precise style direction with vocabulary and references; Refreshing a visual identity — know what's rising vs saturating; Building mood boards — combine trends intentionally with data-backed rationale."
podcast-production
guia-matthieu
"Master Ira Glass's narrative storytelling methodology from This American Life to create podcasts that captivate listeners through the power of story. Use when: Creating a branded podcast or audio content series; Structuring episodes for maximum listener engagement; Transforming raw interviews into compelling narratives; Building thought leadership through audio storytelling; Planning podcast content that keeps audiences coming back"
docker-sandbox
joelhooks
Create, manage, and execute agent tools (claude, codex) inside Docker sandboxes for isolated code execution. Use when running agent loops, spawning tool subprocesses, or any task requiring process isolation. Triggers on "sandbox", "isolated execution", "docker sandbox", "safe agent execution", or when working on agent loop infrastructure.
update-project-doc
flonat
"Update a project's own documentation (CLAUDE.md, README.md, docs/*.md) to reflect its current state. Detects stale file trees, timestamps, counts, and next steps. Includes a leanness audit: flags CLAUDE.md > 200 lines, sections > 15 lines of reference material, SKILL.md > 300 lines, agents > 400 lines, and context files > 200 lines — with auto-extraction to docs/ or references/."
preencher-formulario
dspinillo
EXPERIMENTAL. Preenche no navegador o formulário de candidatura de uma vaga do Radar a partir da folha de respostas, pelo Claude in Chrome, e para antes do envio.
claude-hooks-configuration
laurigates
Set up Claude Code lifecycle hooks and event handlers in settings.json. Use when you want to trigger a script on session start, run a hook before or after tool calls (PreToolUse/PostToolUse), configure hook timeouts to prevent cancellation errors, or debug hooks that aren't firing correctly.
ai-ready
viktor-silakov
Analyzes repositories for AI agent development efficiency. Scores 8 aspects (documentation, architecture, testing, type safety, agent instructions, file structure, context optimization, security) with ASCII dashboards. Use when evaluating AI-readiness, preparing codebases for Claude Code, or improving repository structure for AI-assisted development.
blueprint-promote
laurigates
"Move generated artifact to custom layer to preserve modifications"