Automation
Task automation and scripting
AplicacionMovile
by TheKidin
ogt-docs
by OpenDNDApps
Documentation-as-Source-of-Truth workflow. Use when working with projects that use docs/ as the canonical source for definitions, rules, and tasks. Routes to specialized sub-skills for specific documentation types.
SKILL
by Sunmax0731
文字化け、改行差分、ZIP再生成で揺れる値は固定評価値にしない。
w01-requirement-discovery
by qiao-925
需求发现规范，适用于探索性场景，通过角色扮演帮助用户在迷茫中发现高ROI功能方向。自动触发，纯对话式启发。
hacker-news
by adamelliotfields
Get the latest stories from Hacker News. Use when the user requests the top or newest stories from Hacker News.
test
by eljun
Test web implementations using Playwright MCP for E2E testing. Creates test reports in docs/testing/*.md. Supports email-based auth testing via Mailinator (registration, verification, password reset, notifications).
hreng-1-1-prep
by clous-ai
Use when the user is preparing for a 1:1 meeting and wants an agenda from recent feedback, incidents, goals, and prior notes, with talking points and follow-ups.
visualize-with-libraries
by JefferCB1
Renderiza diagramas BPMN usando iconos de carpetas locales ('icons' o 'librerias'). Detecta automáticamente carpetas de iconos en el proyecto. Genera JSON Excalidraw minimalista con coordenadas X/Y.
linux-ops
by kakahu2015
Agent-native Linux operations skill over SSH. Use when an AI Agent needs to observe, reason about, and operate remote Linux servers using composable primitives. This is an AI-native operations substrate, not a task-script collection: the skill owns safe execution primitives, Python runtime gate enforcement, fail-closed semantic guard, JSON primitive rules, autonomy policy validation, path/risk guardrails, redaction, audit hashes, escalation events, and bounded testable execution; the model owns observation, reasoning, diagnosis, decision-making, verification, and escalation.
executing-plans
by ederheisler
Use when you have a written implementation plan to execute in a separate session with review checkpoints
Continual Learning Guard (EWC / SI / Progressive Networks / Replay)
by sovr610
This skill should be used when the user asks to "prevent catastrophic forgetting", "elastic weight consolidation", "EWC regularization", "progressive networks", "continual learning strategy", "knowledge distillation for retention", "replay buffer memory", "task boundary detection", "fisher information matrix", "synaptic intelligence", "PackNet pruning", "memory-aware synapses", "add continual learning guard", "implement EWC penalty", "add experience replay", "implement progressive columns", "add Fisher diagonal computation", "implement reservoir sampling", "add knowledge distillation loss", "implement task-free continual learning", "add online EWC", "implement generative replay", "add PackNet iterative pruning", "implement synaptic intelligence path integral", or mentions catastrophic forgetting, continual learning, lifelong learning, sequential task training, knowledge retention, task interference, Fisher information regularization, or Phase 8 continual-learning pipeline in the cognitive architecture.
system-create-skill
by maxoreric
v1.0.0 (2026-01-19): åå§çæ¬
webapp-testing
by diyorbekkd
Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.
using-superpowers
by Cygnusfear
Guidelines for using skills effectively - load relevant skills before complex tasks, not every message
workflow-status
by JHaugaard
"Display project workflow progress by reading handoff documents in .docs/ directory. This skill should be used when users want to check their workflow status, see what phase they're in, or when other workflow skills need to verify prerequisites. Provides reusable prerequisite-checking templates for integration with other workflow skills."
plan-review
by rozwer
Review a phase plan interactively, explaining each task and asking for confirmation.
skillcere
by zhangyueyork
"SkillCere control skill for skill recommendation and skill inventory management. Use only when the user explicitly asks to use SkillCere, asks which skills can help with a task, asks for skill recommendations before work, or requests skill management actions such as scan, sync, status, export, or prune. Do not use automatically for ordinary user tasks that do not ask for skill recommendation or SkillCere."
notion-spec-to-implementation
by Victory-Hugo
Turn Notion specs into implementation plans, tasks, and progress tracking; use when implementing PRDs/feature specs and creating Notion plans + tasks from them.
cosmos-delegate-to-agent
by getcosmos
Teaches agents how to effectively use the delegateToAgent tool for orchestrating multi-agent work. Covers dependency graph analysis, parallel vs serial execution patterns, and progress tracking. Use when an agent needs to delegate tasks to specialist agents, especially when working with plans, todos, or multi-step requests.
say-to-her
by bakabird
Transform source material (articles, papers, news, tech content, or any topic) into short, dramatic, entertaining conversational snippets with delivery stage directions. Extracts ONE interesting nugget, dramatizes it with storytelling techniques, and annotates with pauses, voice modulation, gestures, and expressions so even the most introverted engineer can deliver it naturally and engagingly to his girlfriend or partner. Use when: - User wants to turn an article or topic into something to tell their girlfriend/partner - User asks to make technical content conversational or girlfriend-friendly - User wants a casual script for sharing something interesting - User mentions "say to her", "tell her about", "explain to my girlfriend" - User wants dinner conversation material or talking points from source material Triggers: "say to her", "tell my girlfriend", "girlfriend-friendly", "make it conversational", "how do I explain this to her", "turn this into something I can say", "dinner conversation"
chrome-extension
by fanthus
Guides creation and modification of Chrome extensions using Manifest V3. Use when building browser extensions, Chrome plugins, or when the user mentions manifest.json, content scripts, background service worker, popup, or extension permissions.
moltbot-best-practices
by NextFrontierBuilds
Best practices for AI agents to avoid common mistakes. Learn from real failures - confirms before executing, shows drafts before publishing, stops when told to stop, and doesn't over-automate.
checkmate
by InsipidPoint
"Focus your agent on a long-running task and keep iterating until your satisfaction criteria are met — not just a best-effort attempt. Use when you want to lock your agent into a quality loop: it converts your goal into machine-checkable criteria, spins up a worker, judges the output, feeds back gaps, and repeats until PASS or you call it done. Good for tasks where correctness matters: code that must pass tests, docs that must hit a standard, research that must be thorough, anything where 'good enough on first try' isn't acceptable. Triggers on 'checkmate: TASK', 'keep going until it passes', 'don't stop until done', 'quality loop', 'iterate until satisfied', 'judge and retry'."
metaapp-builder
by metaid-developers
Package completed MetaApp frontend projects into distributable zip archives. Use this skill when users request to package, bundle, or create a distribution archive of their MetaApp project. Automatically detects MetaApp project structure, validates required files, and creates a timestamped zip file. Trigger phrases include: "帮我打包xxx项目", "打包这个MetaApp", "创建dist包", "生成zip压缩包".