AI & ML

dengyie Bundle

Use when creating new Codex threads, agents, or dev processes with isolated branch-bound worktrees; fixing Codex UI/thread/worktree coupling; repairing detached HEAD Codex worktrees; or protecting primary main/master worktrees while setting up branch-bound development agents.

facebook Skill

Use this whenever debugging an iOS DAT app with local DAT Inspector MCP tools, live device events, Meta AI app/device boundary issues, permissions, registration, sessions, streaming, callbacks, or user reports that the app cannot communicate with glasses.

fengmo11 Bundle

Use when drafting, auditing, or improving data availability statements, code availability statements, FAIR metadata, dataset access routes, repository choices, restrictions, identifiers, and result-to-data traceability.

shokai Skill

Agentツール（subagent）と相談するスキル。 ユーザーが「subagentと相談して」「subagentに聞いて」「subagentにレビューしてもらって」と言った時に使用する。 現在の会話コンテキストに基づいてsubagentにプロンプトを送り、結果を要約して報告する。

slicervm Skill

Move a git worktree or repository into a Slicer microVM with a working, self-contained .git — prefer agent `--worktree` mode for coding sandboxes, use `slicer wt push` for manual VM flows, then pull commits back.

slicervm Skill

Record two coding agents racing the same task side by side — one agent per fresh Slicer microVM, each bridged into a host xterm (slicer vm shell) on a host Xvfb display, captured with ffmpeg. Use when asked to record a two-agent race, side-by-side model comparison, or dual-terminal demo video.

slicervm Bundle

Record a terminal or TUI (coding agent) demo as a real, watchable video using Xvfb + xterm + ffmpeg x11grab inside a Slicer VM. Use when asked to record a terminal demo, capture an agent run as video, produce a screen recording of a CLI session, or trim/deliver such a recording.

vngcloud Bundle

Platform reference and getting-started guide. Use for general questions about the platform, architecture, available services, SDK, IAM setup, credentials, or which skill to use. IMPORTANT - If the user wants to BUILD an agent (not just learn about the platform), use /agentbase-wizard instead — do not answer with reference material. Trigger phrases include how does the platform work, what services are available, platform overview, explain the architecture, which skill should I use, how do I get started, what can I do here, how do I set up credentials, IAM setup. This is a reference guide — for specific operations, invoke the dedicated skill.

vngcloud Skill

Manage platform LLM model access and API keys for AI agents. Use when user wants to set up LLM access, get a platform API key, browse available models, choose which LLM to use, configure model access, check rate limits, or get an OpenAI-compatible endpoint. Trigger for which models are available, set up LLM, API key for the model, list models, LLM key, connect to LLM. When user says API key without specifying an external service name, default to this skill. DO NOT use for storing API keys for external services (OpenAI, Google, Slack) — use /agentbase-identity instead.

vngcloud Bundle

Add memory to AI agents — conversation history, semantic fact extraction, and long-term memory records. Covers creating memory stores, managing events/sessions, memory records, long-term memory strategies (SEMANTIC, USER_PREFERENCE, CUSTOM), and LangChain/LangGraph memory integration. DO NOT use for agent source code (use /agentbase-wizard), agent identity or outbound auth (use /agentbase-identity), platform LLM API keys (use /agentbase-llm), or runtime logs (use /agentbase-monitor).

vngcloud Bundle

Register agent identities and manage outbound authentication providers for AI agents. Covers CRUD for agent identities, storing API keys or secrets for external services (OpenAI, Google, Slack), OAuth2 providers, and delegated keys. When user mentions an external service name with API key or credentials, trigger this skill — not /agentbase-llm. DO NOT use for agent source code (use /agentbase-wizard), platform LLM API keys (use /agentbase-llm), agent memory (use /agentbase-memory), or runtime logs (use /agentbase-monitor).

wesleysimplicio Skill

Image, video, music, 3D, audio, and LLM generation via PiAPI. Wraps the unified `POST /api/v1/task` envelope plus the OpenAI-compatible `POST /v1/chat/completions` LLM proxy.

wesleysimplicio Skill

Use this skill when the user asks for image, video, music, 3D, audio, or LLM generation backed by the PiAPI inference platform (https://piapi.ai). Covers Midjourney, Suno, Kling, Luma, Hailuo, Flux, Hunyuan, Faceswap, Trellis, MMAudio, F5-TTS, Veo 3, Gemini 2.5 Flash Image, and Seedance via a single unified `POST /api/v1/task` envelope. The CLI also wraps the OpenAI- compatible `POST /v1/chat/completions` LLM surface.

wesleysimplicio Skill

PiAPI inference gateway — image, video, music, 3D, audio, and LLM generation through one async-task envelope and an OpenAI-compatible LLM proxy.

wesleysimplicio Skill

Generate images, video, music, 3D, audio, and chat completions through PiAPI's unified async-task gateway. Aggregates Midjourney, Flux, Kling, Luma, Hailuo, Veo 3, Suno, Hunyuan, MMAudio, F5-TTS, Trellis, Gemini Nano Banana, Seedance, and Faceswap behind a single `POST /api/v1/task` envelope, plus an OpenAI-compatible `POST /v1/chat/completions` LLM proxy.

wesleysimplicio Skill

Portable, agent-agnostic spec for the PiAPI inference gateway. Single async-task envelope at `POST /api/v1/task` aggregates Midjourney, Flux, Kling, Luma, Hailuo, Veo 3, Suno, Hunyuan, MMAudio, F5-TTS, Trellis, Gemini Nano Banana, Seedance, and Faceswap. `POST /v1/chat/completions` is an OpenAI-compatible LLM proxy. Auth header is `x-api-key`.

wesleysimplicio Skill

PiAPI inference gateway. One unified `POST /api/v1/task` envelope for Midjourney, Flux, Kling, Luma, Hailuo, Veo 3, Suno, Hunyuan, MMAudio, F5-TTS, Trellis, Gemini Nano Banana, Seedance, and Faceswap. Plus an OpenAI-compatible `POST /v1/chat/completions` LLM proxy.

wesleysimplicio Skill

PiAPI inference gateway. One async-task envelope plus an OpenAI-compatible LLM proxy. Aggregates Midjourney, Flux, Kling, Luma, Hailuo, Veo 3, Suno, Hunyuan, MMAudio, F5-TTS, Trellis, Gemini Nano Banana, Seedance, and Faceswap.

eyriecommander Bundle

Use when the user asks how to do a specialized task, asks whether a skill/tool exists, wants to extend agent capabilities, or a task looks like a repeated workflow that might benefit from an existing installable skill.

eyriecommander Bundle

Rewrite or review prose to remove formulaic LLM tells while preserving meaning, factual claims, citations, privacy boundaries, and the user's intended voice. Use when the user asks to humanize, de-AI, de-slop, make text sound more natural, make prose less generic, match a voice sample, polish public copy, or audit writing for AI-ish patterns. Do not use for deception, citation laundering, factual invention, or bypassing academic, legal, employment, or platform rules.

devcsde Skill

Check OATDA account balance, remaining credits, and total usage via the unified AI gateway. Triggers when the user asks "how much OATDA credit do I have left", "is my budget enough for this AI call", "what's my OATDA usage", "am I out of credits", "verify balance before image/video/LLM generation". Returns current balance, usage spent, and available remaining amount. Useful for budget checks before expensive image, video, or long-running LLM calls.

devcsde Skill

Compare multiple LLM models (GPT-5, Claude Sonnet 4.5, Gemini 3, DeepSeek v4-pro, Grok 4, etc.) side-by-side in one parallel request via OATDA's unified AI gateway. Triggers when the user wants to benchmark models, evaluate quality differences, pick the best LLM for a task, run "gpt vs claude vs gemini" comparisons, do multi-model evaluation, or find the cheapest/best/fastest answer. Returns each model's output plus token usage and cost per model.

devcsde Skill

Generate text, code, or chat completions using OATDA's unified LLM API gateway — like calling ChatGPT, Claude, Gemini, GPT-5, Claude Sonnet 4.5, Grok 4, or Qwen from a single endpoint. Triggers when the user wants to generate, write, summarize, translate, or complete text via OpenAI, Anthropic, Google, Deepseek, Mistral, xAI, Alibaba/Qwen, MiniMax, ZAI/GLM, Moonshot/Kimi, or Bytedance. Also use when comparing outputs across models or using aliases like "gpt-4o", "gpt-5", "claude", "gemini", "deepseek", "grok", "qwen", "mistral", "kimi".

ilonatommy Bundle

A meta-skill for managing Agent Skills. Use when creating, editing, modifying, updating, deleting, or refactoring skills. Also triggers for SKILL.md files, ~/.copilot/skills/ directory, .github/skills/ folder, skill frontmatter, skill instructions, skill templates, or any file changes in a skills directory. Ensures all changes are committed and pushed to git.

edonadei Bundle

Measure a skill's reliability — run it k times for a pass@k score, design or interpret its eval, or compare it against the base agent. Use when the user wants to run, design, or interpret a skill's eval, or write an .eval.yaml spec.

edonadei Bundle

Convert a title into a URL slug. Use when asked to slugify a title or make a URL slug.

m1nga Bundle

Turn rough product ideas, conversations, documents, or an existing codebase into an end-to-end product system map covering user journeys, capability boundaries, platform and customer mandates, data lifecycles, agent/service/human responsibilities, architecture, failure paths, governance, and delivery slices. Use when the user asks to understand the whole product, find missing functions or branches, establish product foundations, decide what belongs in code versus mandates or configuration, design multi-agent responsibilities, audit product completeness, or convert a product concept into an executable architecture. Trigger on phrases such as “看一下产品结构”, “梳理产品架构”, “检查用户旅程和功能分支”, “review the product structure”, or “map the whole product”.

mattmireles Skill

Audit Gemma Multimodal Fine-Tuner for training correctness, dataset leakage, MPS/device behavior, checkpoint/export integrity, evaluation validity, privacy, and CLI/wizard/visualizer regressions.

mattmireles Skill

Edit Gemma Tuner Markdown while preserving links, runnable commands, tables, heading structure, and honest model/data/hardware claims.

haaaiawd Bundle

综合 Scout 阶段所有分析（build-inspector, runtime-inspector, git-forensics, concept-modeler），生成决策就绪的系统风险报告。

netclaw-dev Bundle

REQUIRED when the user asks about scheduling, reminders, cron jobs, timers, background jobs, diagnostics, troubleshooting, MCP tools, daemon health, identity updates, or Netclaw capabilities and self-maintenance.

chloechen16788 Bundle

官方Viking CLI 命令行助手：本CLI覆盖火山引擎/BytePlus VikingDB(向量库)/Knowledge(知识库)/Memory(记忆库)的数据集管理和数据的读写及检索， 可用于扩展Agent的知识检索边界 提升Agent的记忆能力； 当用户对知识库/向量库提问时，使用本Skill; 当用要操作向量库/知识库 或 从向量库/知识库检索信息时使用本Skill； 当用户要记忆检索和记忆存储时，使用本Skill。

chloechen16788 Bundle

Guide a user conversationally through generating a new ad video from a template video and a user-provided product image. The host agent performs video understanding, detailed analysis, and rewrite planning; BytePlus Seedance is used only for final video generation with local product images as generation references.

stacksjs Bundle

Use when writing or editing any document an agent reads - a SKILL.md under app/Skills or storage/framework/defaults/ai/skills, the project AGENTS.md, or a reference file a skill points at. Covers context pointers, the information hierarchy, completion criteria, leading words, pruning, and the skill mechanics behind app/Skills and buddy setup:ai.

greatmark Skill

Surface the user's unknown unknowns before they start work in an unfamiliar area (new codebase area, domain, or tool). Produces a read-only blindspot briefing plus advice on how to prompt better. Use when the user says "blindspot pass", "盲区扫描", "unknown unknowns", asks what they don't know or might miss before diving in, says "我刚接手这块代码", "开工前帮我看看有什么坑", or merely describes facing unfamiliar territory — the statement is the request. Not for finished or pre-merge work (use change-quiz), nor for interviewing the user (use interview-me); mixed requests start here — brief, then hand off.

nikhil8bph Bundle

Create and maintain the UX design handoff for this product from approved requirements, architecture, and API contracts using the Google Stitch MCP. Produces the canonical `docs/DESIGN.md` consumed by the Angular implementation skill; do not use for application code or contract changes.