AI & ML

bitsky-tech Bundle

Build agents with the Bridgic Amphibious dual-mode framework — combining LLM-driven (agent) and deterministic (workflow) execution with peer state-machine dispatch, step-level fallback (a bounded agent recovery sub-run), human-in-the-loop, and a two-loop context model: a small-loop OTAContext (framework-owned: this run's user_input + observe-think-act round trace + tools) and a free-form big-loop Context (cross-turn knowledge). Tools are declared on the OTA context via OTAContext.tool — nothing is auto-injected. Sub-goals can be delegated to external coding agents (claude code, codex) as think-agent units. Use when: (1) writing code that imports from bridgic.amphibious, (2) creating AmphibiousAutoma[OTAContext, Context] subclasses, (3) defining CognitiveWorker think units (think_unit / ThinkUnit) and AgentWorker think agents (think_agent / ThinkAgent), yielding ThinkUnit / ThinkAgent / EnterAgent / ActionCall / HumanCall / LLMCall / RETURN, (4) implementing on_agent / on_workflow methods, (5) working with the

coplaydev Bundle

Orchestrate Unity Editor via MCP (Model Context Protocol) tools and resources. Use when working with Unity projects through MCP for Unity - creating/modifying GameObjects, editing scripts, managing scenes, running tests, or any Unity Editor automation. Provides best practices, tool schemas, and workflow patterns for effective Unity-MCP integration.

hypier Bundle

Use when current TradingView data must be retrieved through available MCP tools for quantitative market analysis, including screening, technical analysis, risk assessment, event analysis, market review, or multi-symbol comparisons.

kiluazen Bundle

Use when an agent needs to operate the user's real Chrome session — listing tabs, snapshotting the page into actionable @refs, clicking, filling, typing into rich editors, pressing keys, evaluating JS, capturing screenshots, and reading console/network buffers. All actions go through CDP and run on backgrounded tabs without stealing focus.

meleantonio Skill

Build a CDC-style problem-specific prompt for attacking an open math problem: exact restatement, complete success criteria, non-counting partial results, traps, multi-agent search heuristics, and adversarial audit requirements. Use when creating PROMPT.md, generating an agent goal prompt, or invoking /craft-proof-prompt. Keywords: proof prompt, CDC prompt, problem-specific prompt, Erdős prompt.

microsoft Skill

Build MCP (Model Context Protocol) clients to connect AI agents with external tools and data sources. Use when building integrations with MCP servers.

yajnignez Bundle

Reverse-engineer an uploaded or referenced music file into one precise Suno-ready English generation prompt under 850 characters. Use for audio-to-prompt analysis of melody, harmony, rhythm, timbre, vocals, arrangement, effects, mix, and mastering. Avoid creator and work names, broad genre triggers, and preset-like imitation; express the track through ranked, audible production evidence.

yamadashy Bundle

Persistent file-based notebook (.carnet/<category>/<slug>.md, auto-expiring after 30 days of disuse) for AI agents and humans. Use whenever saving, recalling, or organizing notes across sessions — IMPLICIT cues include the user recapping a tricky fix worth not redoing, asking 'what did we decide about X', handing off mid-task state, or setting something aside for now. Use PROACTIVELY when you discover something a future agent would otherwise re-derive from scratch — a non-obvious library gotcha, a hard-won fix, a rejected design alternative likely to be re-proposed, an in-progress state worth resuming. SKIP for: time-bound TODOs that need to nag (TodoWrite/Issues surface back, carnet does not), credentials/secrets (don't put them on disk), context already in a commit/PR/ADR (no double-bookkeeping). When in doubt, prefer a brief note over losing context: unused notes expire automatically.

yezhwi Bundle

Distill an open-source framework into reusable architecture knowledge. Use when reverse engineering a framework, runtime, workflow engine, orchestration platform, or plugin system to reproduce its architecture, execution model, extension points, and engineering tradeoffs without copying implementation.

yl86319749-pixel Bundle

医疗器械与体外诊断试剂注册申报资料编写助手，调用远程 MCP 工具自动检索分类目录和国家标准，生成符合 NMPA 格式的注册资料章节。

yogsoth-ai Skill

Spawn Agent

ytubecoder Skill

All-in-one skill — setup whisper.cpp, launch the capture app, analyze a session, or watch a live capture via MCP. Works from any project.

yukino1338 Bundle

Conservatively clean local Chinese plain-text (.txt) novels by preserving an immutable source snapshot, fully reviewing external advertisements or watermarks, compiling and atomically applying exact-anchor decisions, verifying the final layout, exporting traceable TXT, Markdown, and EPUB outputs, and optionally reporting suspicious chapter-title or masked-word candidates without changing them. Use when an Agent is asked to clean, purify, remove ads or watermarks from, inspect title or masking anomalies in, or safely export a Chinese TXT novel; do not use for general document editing, semantic rewriting, or ebook-library management.

z2z23n0 Bundle

Use when an agent needs to install or use Agent Capsule to export, share, inspect, import, restore, or verify Codex and Claude Code sessions, or to migrate a Codex profile between Macs. Covers session capsules, encrypted links, target-device orchestration, project selection and cloning, controlled profile overwrite, one-shot offline import, and verification.

zaimokuza-yoshiteru Skill

Generate or refine an Agent Skill for an MCP server. Use when the user provides an MCP Profile, asks to expose MCP tools through an agent-readable CLI workflow, or needs a verified SKILL.md with safe invocation rules.

zapier Bundle

Install and connect Zapier and route the person to the right surface (MCP, SDK, or CLI). Use for typed or spoken requests to install, add, connect, or set up Zapier (heard over voice as "zap ear", "zappier", or "zap year"), when someone wants an AI client (Claude, ChatGPT, Cursor) to use their apps, wants to call Zapier from code, or wants it in the terminal, or when they describe the intent without naming Zapier, such as "connect my apps" or "hook up my tools".

zeropointrepo Skill

YouTube transcripts, video and channel search, channel browsing and playlist extraction via the bundled TranscriptAPI MCP server. Use when YouTube is or could be relevant — pasted video, channel or playlist links, video IDs, @handles, requests to summarize, quote, transcribe, translate or fact-check a video, research a topic through video, or browse what a creator has posted. Not for uploads, comments or account management.

zhang2023-byte Bundle

Build and use a small, independent, reusable scientific-data harness in explicit stages. First confirm the scientific contract and construct offline tooling; then let the user provide dates and confirm agent-recommended arXiv queries before archiving papers; finally run isolated one-paper extractions concurrently, save valid evidence results, show the resulting dataset, and close with practical extension paths. Use for building traceable, multi-quantity, AI-ready datasets from scientific literature without generating classroom experiment configurations.

zhuyansen Bundle

Use when the user wants to find, evaluate, audit, or install an open-source AI agent skill or MCP server — e.g. "find an MCP server for Postgres", "is this skill safe to install", "what should I use to scrape a website". Searches a quality-scored, security-graded catalog of ~20K skills locally (cached index, zero backend load) and returns each result's security grade, quality score, and install commands so you can check trust BEFORE installing.

zhwenchao Bundle

Consent-first profile drafting, confirmation, and manual correction interface for Hermes Agent-integrated Jungian Soul Agent skills. Provides versioned local state layer with explicit user approval before persistence.

zjp1997720 Bundle

Audit Codex installation health and workspace context quality by combining the built-in `codex doctor` report with read-only checks for AGENTS.md scope, instruction bloat or duplication, Skills, MCP, hooks, config, Git state, and repository-root hygiene. Use whenever the user asks for `/doctor`, `/checkup`, Codex health checks, context cleanup, AGENTS.md cleanup, unused or broken Skills/MCP/hooks, slow or polluted context, or asks why Codex is ignoring rules. Default to diagnosis only; never edit, disable, delete, trust, install, update, authenticate, move, or clean anything without a separately approved finding-level diff.

zjp1997720 Bundle

在 Codex App 中为复杂、可并行的知识工作或编程任务自动创建多个可指定模型与推理强度的后台任务，由主 Agent 负责规划、分工、集成和验收。用于多来源调研、多章节内容、复杂 Skill/PPT、跨模块开发、独立验证或 2 个以上互不依赖工作流；也用于用户明确要求模型路由、后台 Worker、Agents Team 或持久项目协作。简单问答、状态查询、单文件小改、强顺序任务和发布/付款/删除/账户操作不得自动触发。

zjp1997720 Bundle

Package and publish one local agent Skill, or audit and release a Registry-driven public Skill portfolio, as polished installable GitHub packages. Use when the user gives a SKILL.md, Skill directory, or multi-Skill repository and asks to open-source, publish, release, share, validate, detect drift, synchronize mirrors, manage versions, make installable with npx skills, prepare release documentation, or generate launch copy. The workflow includes safety, self-containment, declared capabilities, project-native README design, isolated installation, release planning, and publishing.

zmgid Bundle

电商商品图技能，也能当普通生图工具用。套图走模板（一个文件夹：template.json + 示例图），三种做法：replace（甲方样图换货，脚本直出）、smart（模板给每槽 brief，按品写 prompt）、design（没模板，从零问需求设计一套）；同一甲方一大夹混了多个大类走甲方大单（先 要求.json，按大类分夹，一类一模板，每类先出 2 个再铺）；零散一张几张图走 gen。出图走 OpenAI / Grok / Gemini 或兼容网关的图片 API。Use when the user says 使用 dsimage / dsimage / dsimage引导 / 打印dsimage引导, or asks 怎么用 / 不会用 / 使用说明, or asks for 电商主图 / 详情页 / 产品图 / 商品图 / 白底图 / 套图 / listing images / product photos / PDP / A+ content, or 生成一张图 / 出张图 / 改图 / 换背景 / 海报 / generate an image / edit this image, or 制作模板 / 创建模板 / 换货 / 换品 / 替换模板 / 按这套样图做, or 甲方大单 / 一个大文件夹很多品 / 按大类做模板.

zq940222 Bundle

Create professional 3D designs in Blender via the official Blender MCP server. Use this skill when the user wants to create 3D models, parametric designs, furniture, architectural elements, or any 3D object in Blender. Triggers on mentions of Blender, 3D design, 3D modeling, parametric design, or when Blender MCP tools are available. Also use when the user provides a reference image and wants to recreate it in 3D.

zxyasfas Bundle

Use this skill when a user needs to inspect, score, repair, or regression-test academic paper formatting in DOCX files, especially thesis, journal, or conference documents with a separate format specification. The skill guides Codex through local-first document handling, rule extraction, formatting repair, report review, and contribution-safe validation.

zzmklb0906-prog Bundle

Plan and review software changes while delegating implementation and test execution to AGY through the agy-worker-mcp-runtime MCP tools. Use when the user requests the Codex-lead and AGY-worker engineering workflow. Do not use for ordinary coding tasks that do not request AGY delegation.

6551team Bundle

Twitter/X data via the 6551 API. Supports user profiles, tweet search, user tweets, follower events, deleted tweets, and KOL followers.

grey-iris Skill

Use this skill when a user wants to turn meeting notes into a tracked Action Items database in Notion, or to bulk-edit, find-replace, normalize, or repair Notion database rows or page text at a scale past Notion's native find-and-replace limits. Uses easy-notion's MCP tools.

wanbinyu Bundle

Decide whether to use an Agent Skill, an MCP server, both, or neither. Use when choosing architecture for agent capabilities, or Chinese "skill 还是 MCP", "用 skill 还是 tool", "要不要做 MCP". Decision skill - not an implementation tutorial.

yinxiaoyu-1998 Bundle

上传和查询企业数据中枢（SMEDC）中的文件与资料：当员工要求把文件/文档/表格上传到 企业数据中枢、企业资料中枢、企业知识库或企业数据库，或想基于已上传的企业资料提问业务问题时使用。 用户要求更新 smedc-mcp 或 smedc-mcp-skill 这个 skill 时也使用本技能。 用户询问 SMEDC 的配套分析、报表或扩展 skill 时也使用本技能。 Also use when users ask about optional SMEDC analysis, reporting, or extension skills. Also covers updating this skill from its official GitHub source and installing, configuring, and authenticating the official SMEDC remote MCP launcher (api.smedatacenter.xyz); accesses SMEDC through the launcher only and never exposes employee credentials or service infrastructure.

adaptyteam Skill

Adapty CLI setup

agentlyhq Skill

Discover and communicate with thousands of AI agents, tools, data sources, and APIs on the Agently AI Agent Marketplace/Directory. Use this skill when browsing available agents, sending messages via the A2A protocol, calling MCP tools, making HTTP/Web requests, or interacting with paid agents using automatic x402 micropayments. ALL protocol interactions (MCP, Web/HTTP, A2A) MUST go through the use-agently CLI.

aktsmm Bundle

Design, review, and improve agent workflows & agent using SSOT, SRP, Fail Fast principles. Supports Prompt Chaining, Parallelization, Orchestrator-Workers patterns.

apache Skill

Build the SkyWalking Java agent — full build, skip tests, single module, or plugin test scenarios

ariobarin Bundle

Inspect current LLM tradeoffs across quality, price, speed, context, modality, and OpenRouter availability. Use for model recommendations, comparisons, pricing checks, and code changes that add or swap an LLM.