AI & ML

tomevault-io Bundle

Use when asked to launch/monitor/verify/stop a multi-agent workspace managed by Tutti.

tomevault-io Bundle

Guide NeMo Speech users through ASR fine-tuning with container setup and Lhotse training. Use when this capability is needed.

tomevault-io Bundle

LLM Wiki Manager

tomevault-io Skill

Your AI CoPilot on Mobile — or give your AI its own phone. Make calls, send SMS, speak via TTS on speakerphone, automate UI, manage files, search media, and 40+ more tools via MCP. Open source, self-hosted, privacy-first. Use when this capability is needed.

tomevault-io Bundle

Help the user create a new zot extension (slash command, LLM tool, or guard) in any language. Use when this capability is needed.

tomevault-io Bundle

Guide for developing with Prism PHP package - a Laravel package for integrating LLMs. Activate or use when working with Prism features including text generation, structured output, embeddings, image generation, audio processing, streaming, tools/function calling, or any LLM provider integration (OpenAI, Anthropic, Gemini, Mistral, Groq, XAI, DeepSeek, OpenRouter, Ollama, VoyageAI, ElevenLabs). Activate for any Prism-related development tasks. Use when this capability is needed.

tomevault-io Bundle

Assesses team fitness and composes agent teams. Use when "set up a team", "team for this", "should I use agents", "design a team", "how many agents", "agent team". Use when this capability is needed.

tomevault-io Bundle

使用 fscan 进行网络扫描和安全评估。当用户要求扫描网段、探测主机存活、发现开放端口、识别服务、检测漏洞或弱口令时使用。支持 NDJSON 结构化输出，适合 AI agent 管道消费。 Use when this capability is needed.

tomevault-io Bundle

Use when creating, modifying, or documenting TypeSpec domain models. Triggers include adding new entities, value objects, enums, extending base types, or when asked to create a "tsp model", "domain model", "entity", or work with files in the tsp/ directory. Part of the Shep autonomous SDLC platform — https://shep.bot

tomevault-io Bundle

Teaches how to interact with the Ray application. This skill should be used when users want to interact with Ray through a coding agent or LLM with skills capabilities. Use when this capability is needed.

tomevault-io Bundle

Shareable TSPlay Flow authoring skill for coders. Use when Codex needs to write, modify, fix, review, or explain TSPlay `.flow.yaml` files, choose between Flow DSL, Lua script mode, and MCP workflows, draft or review tutorial flows, send email notifications with `send_email`, attach to real Chrome/Chromium over CDP, use `browser.cdp_launch`, or debug artifacts and saved sessions. Typical English triggers include write flow, fix flow, review flow, generate flow, repair selector, attach Chrome over CDP, use cdp_launch, send email flow, email notification flow, and convert a requirement into Flow. Typical Chinese triggers include 编写 Flow, 写 Flow, 改 Flow, 修 Flow, 生成 flow, 补一条 flow, 调试 flow, 排查 flow, 设计教程 flow, 接管 Chrome, cdp_launch, 发邮件 flow, 邮件通知 flow, and 把需求转成 TSPlay Flow. Use when this capability is needed.

tomevault-io Bundle

Systematically evaluate your LLM application with TruLens Use when this capability is needed.

tomevault-io Bundle

Build or update MCP server/client integrations for VMark. Use when configuring MCP servers, adding MCP tools, or updating MCP-related docs and settings. Use when this capability is needed.

tomevault-io Bundle

多引擎 AI 创作工具链。TTS 语音合成（edge-tts / Qwen3-TTS）、ASR 语音识别与卡拉OK字幕、ComfyUI 图片生成（ERNIE/Qwen/Z-Image + PromptKG 知识图谱）、视频剪辑（Cut）。使用场景：(1) 文本转语音播放，(2) 音频转录生成 SRT/ASS 字幕，(3) AI 图片生成与风格探索，(4) Prompt 知识图谱查询与模板发现，(5) 字幕级视频剪辑。触发词：语音、TTS、ASR、字幕、图片生成、prompt、知识图谱、KG、视频剪辑、cut Use when this capability is needed.

tomevault-io Bundle

Use when routing Claude Code through a local LiteLLM proxy to GitHub Copilot, reducing direct Anthropic spend, configuring ANTHROPIC_BASE_URL or ANTHROPIC_MODEL overrides, or troubleshooting Copilot proxy setup failures such as model-not-found, no localhost traffic, or GitHub 401/403 auth errors.

tomevault-io Bundle

读取指定文档，生成结构化摘要。作为 compiler 的子 Agent 被调用。 Use when this capability is needed.

tomevault-io Bundle

Aidemd Mcp Server Aide

tomevault-io Bundle

Use when the user wants to generate, edit, or extend arksim test scenarios. Reads the agent's source code to derive realistic scenarios; can build regression scenarios from past failures.

tomevault-io Bundle

Quick inline check for documentation freshness. Use when you've just made code changes and want to know if any docs need updating, without a full audit. Lighter weight than the /documentation agent. Use when this capability is needed.

tomevault-io Bundle

LLM-powered injection of project context into installed agent templates via `aspens customize agents` Use when this capability is needed.

kylezantos Bundle

Fable → Opus delegation. In a Fable session, the main session delegates implementation to Opus subagents and acts as the orchestrator and quality gate, saving Fable tokens for judgment work. Use when the user says "delegate", "use Opus subagents", "orchestrator mode", or wants to conserve Fable budget on implementation-heavy tasks. Only applies when the session model is Fable — Opus delegating to Opus saves nothing.

lewislulu Bundle

Build and maintain a Karpathy-style LLM knowledge base — a self-compiling Obsidian markdown wiki where an Agent ingests raw sources, compiles cross-linked concept/entity/summary pages, answers queries against the corpus, lints the graph for health, and audits in-context human feedback filed from Obsidian or the local web viewer. Use when (1) scaffolding a new knowledge base for any research topic, (2) ingesting articles/papers/PDFs/web pages into raw/, (3) compiling or restructuring wiki articles from existing raw material, (4) answering questions against the wiki and filing durable answers back, (5) running lint passes for dead links / orphan pages / coverage gaps / audit shape, (6) processing human feedback from the audit/ directory and applying corrections. Not for general note-taking, daily journals, or non-wiki Obsidian use.

liangdabiao Bundle

带货视频反向拆解与 Seedance（即梦）复刻提示词生成。当用户提供带货/电商/广告/商销/种草/千川视频（或提到即梦、Seedance、文生视频、复刻爆款、拆解视频、提示词），想分析镜头、复刻效果、生成视频提示词时，使用本skill。自动产出镜头级分镜表（景别/运镜/光影/动作时序/卖点）和符合 S-A-C-S-C 编导规范的完整提示词包（分镜头提示词+基础参数+负面词+@参考图清单），支持忠实复刻/自然流/付费流三种风格。

liangdabiao Bundle

Flue Agent Framework v0.9.1+ — TypeScript harness-driven agent framework. Covers createAgent, init, defineTool, defineAgentProfile, routing, SSE streaming, production deployment, and all known pitfalls.

liangdabiao Bundle

生成 AI 短视频（Seedance 2.0 / 即梦 / 小云雀）分镜提示词的专家。当用户想生成任何形式的短视频提示词、视频脚本、分镜脚本时使用——包括但不限于：剧情/叙事视频、打斗/动作视频、风景旅拍、短剧、伪纪录片、家装空间漫游、UI 产品宣传片、口播视频、音乐卡点、品牌广告片 TVC、产品展示、电商带货视频（抖音/TikTok）、真人感种草/探店视频、视频延长或改剧情。只要用户提到"生成视频提示词"、"帮我写分镜"、"做个短视频"、"带货视频"、"产品广告"等就应触发。

liangdabiao Bundle

AI驱动的电商评论深度分析工具，支持22维度智能标签、用户画像识别、VOC洞察和可视化看板生成。 当用户需要以下功能时触发： - 分析电商产品评论（Amazon/eBay/AliExpress等平台） - 从评论中提取用户画像、痛点和VOC（客户之声） - 生成产品洞察报告和机会点分析 - 创建专业的可视化分析看板 - 进行竞品分析和市场定位研究 触发关键词：电商评论分析、评论分析、竞品分析、用户洞察、VOC分析、产品优化、市场调研、评论数据挖掘 AI Agent 约束：必须通过 AskUserQuestion 收集分析数量后再执行分析

liangdabiao Bundle

Comprehensive weekend city travel investigation skill for Chinese cities. Use this skill PROACTIVELY whenever the user wants to research a city for weekend or near-future (within 1 month) travel — including 小红书 activities, 演唱会/concerts, 集市/markets, 球赛/sports matches, 博物馆/museums, 优惠门票/discount tickets, 喜茶门店/Heytea locations, 美食街/food streets, city walk routes, 5A 景区/scenic areas, and 地铁路线/subway routes. Triggers on phrases like "调研XX城市"、"XX城市周末去哪"、"周末小旅游"、"XX城市旅游攻略"、"本周末/下周末去XX"、"XX城市近期活动"、"weekend trip to [city]"、"investigate [city]"、"城市调查". Search via anysearch CLI. All judgment, extraction, and writing done by the runtime agent (you) — no external LLM API calls. Generates text-only Markdown reports (no inline images; anysearch `search` does not return thumbnailUrl). Always invoke this skill when the user mentions researching any Chinese city for short-term travel, even if they don't explicitly ask for a "skill" or "攻略".

lidge-jun Skill

Activate an xAI OAuth session as a local Grok proxy and CLI tool surface. Use when you need Grok models (chat, reasoning, search, images, video, voice, RAG) from OpenAI-compatible clients or agent tools. Requires a SuperGrok subscription + `progrok login` once.

lltx Bundle

Use this skill when diagnosing Java applications with Arthas through the arthas-mcp-server, including thread analysis, hot methods, exceptions, class inspection, method tracing, watch/stack/monitor workflows, or when the user asks to use Arthas or Arthas MCP. This skill connects to a fixed streamable HTTP MCP endpoint at http://localhost:8563/mcp.

lobehub Skill

将本仓库的 Go gateway 实现与上游 LobeHub 参考实现（device-gateway / agent-gateway）以及应用侧客户端协议进行对齐。当需要同步上游协议、追平最新实现、或对齐到指定 LobeHub tag/commit 时调用。

lucianwhy Bundle

跨Agent全局记忆库管理工具。支持 remember（写入）、recall（查询）、archive（归档洞察）、hygiene（健康检查）、upgrade（自我更新）五个指令。所有记忆以 Markdown 文件存储，人类可直接编辑。

madappgang Bundle

Drive tmux from the command line to manage terminal sessions, windows, and panes — split the current pane, run a command in a new pane, capture output, send keystrokes, resize/navigate panes, or spin up a fully isolated background tmux server for sandboxed work. Use this whenever you need to run something in a terminal pane beside your work, watch a long-running process (dev server, build, test watcher, log tail), drive an interactive CLI/REPL, set up a multi-pane layout, or run commands in a tmux server that stays invisible to the user's own `tmux ls`. Trigger on any mention of tmux, splitting/managing panes or windows, "run this in a pane", "watch the output", side-by-side terminals, or background/isolated terminal sessions — even when tmux isn't named explicitly but the task clearly needs terminal-pane control. If the tmux-mcp tools (`open-pane`, `execute-command`, `start-and-watch`, `capture-pane`, …) are available, prefer them for the ordinary "give me a pane and run something in it" case and use this sk

manaflow-ai Bundle

Use when developing cmux-home, explaining its cmux event-driven Rust TUI, or adapting it into a custom cmux launcher for parallel agent workflows across worktrees, multiple checkouts, SSH hosts, VMs, browsers, or review dashboards.

manavarya09 Skill

Cursor-style checkpoint & undo for Claude Code. Instantly revert to any previous state. Trigger: '/checkpoint', '/cp', 'undo changes', 'revert changes', 'go back to before', 'restore previous version', 'undo last prompt'.

manavarya09 Skill

Use when the user or agent needs to pick a public API for a task (weather, images, finance, etc.). Queries public-apis-live for reachable, deduped options.

manavarya09 Skill

TokenForge — Full-stack LLM token optimization engine