AI & ML

yht0827 Bundle

Open Design 로컬 앱으로 웹/모바일 프로토타입, 대시보드, 랜딩, 덱 산출물을 만들거나 설치/실행/agent 감지/Codex 연결을 점검해야 할 때 쓰는 스킬

yht0827 Bundle

Use when code-review-graph, CRG, graph-based review context, blast-radius analysis, Codex MCP setup, or structural code graph commands are needed for a repository.

jamedwards1-gpt Skill

Develops SAP BTP ABAP (steampunk) objects through ADT MCP with read and high-level write tools, then refreshes the local mirror. Use for ABAP, RAP, CDS, classes, steampunk changes, ADT MCP, or syncing package ZZSD.

mahir-isikli Bundle

Yorck Cinema (Berlin) — Agent Skill

funkeomolere Bundle

Use this skill when the user is preparing for, scoping, or assessing against the EU AI Act (Regulation EU 2024/1689) as a provider, deployer, importer, or distributor of AI systems. Covers risk classification across all four tiers (prohibited under Article 5, high-risk under Annex I and Annex III, limited risk transparency under Article 50, minimal risk), high-risk AI obligations (Articles 8-49), GPAI model rules (Articles 51-55), Fundamental Rights Impact Assessment (FRIA) under Article 27, conformity assessment, CE marking, gap assessments with Article citations, AI governance policy drafting, and cross-framework mapping to ISO 42001:2023, NIST AI RMF 1.0, GDPR, and DORA.

bruhxax Skill

Use whenever the user gives detailed instructions, specific colors, reference images, design requirements, UI requirements, visual style directions, or asks to copy or match a design.

pigbiglong Bundle

把 Hermes Agent 变成你的小红书运营助手 — 首页分析、账号诊断、选题灵感、内容创作、自动发布、评论回复、爆款复刻、知识库沉淀。全链路 SOP 覆盖，基于浏览器自动化（CDP）。

yolo-labs-hq Skill

Run vitest with the project's own config. Prefer `vitest run path/to/file.test.ts` over watch mode during agent sessions to avoid leaving processes behind. Use `--reporter=basic` for compact output the agent can parse, and `--coverage` only when explicitly asked.

yolo-labs-hq Skill

Diagnose an MCP server that is failing to register tools or returning errors. Walks through stdio handshake validation, env-var injection, tool-schema introspection, and host-side debugging so you can localize whether the bug is in the server, the transport, or the host's tool routing.

yolo-labs-hq Bundle

Conducts multi-axis code review. Use before merging any change. Use when reviewing code written by yourself, another agent, or a human. Use when you need to assess code quality across multiple dimensions before it enters the main branch.

kaustubhhiware Skill

Behavioral guidelines to reduce common LLM coding mistakes. Emphasizes thinking before coding, simplicity, surgical changes, and goal-driven execution. Load this when writing or modifying code to avoid overengineering and unnecessary changes.

kaustubhhiware Skill

Prevent repeated agent mistakes across code, docs, tests, and configs by enforcing repository mistake memory before edits and recording new failures.

picard1701ent Bundle

Run and monitor LLM benchmark matrices with a smoke-to-full workflow. Use for multi-model, multi-benchmark execution with tmux/nohup, structured logs, and resumable operations.

picard1701ent Bundle

Validate expensive or long-running LLM pipelines before launching the full run. Use when a workflow has multiple stages such as data generation, training, evaluation, or checkpoint selection and failures would waste significant time, compute, or API cost.

picard1701ent Bundle

Run tests, smoke checks, training jobs, and benchmark pipelines when sandbox, network, GPU, package, or environment issues may block normal execution. Use when a task involves pytest/unittest, experiment scripts, conda environments, API-backed evaluation, model downloads, CUDA, or other commands that may need escalated permissions or a different runtime than the default shell.

woosook0127 Skill

Use when Codex must inspect the repository before architecture, refactor, experiment, automation, or model-work decisions; maps entry points, configs, datasets, engines, metrics, checkpoints, and reuse boundaries with file evidence.

woosook0127 Skill

Use when an idea, paper, model change, dataset, or experiment must be mapped onto the current repository backend, launcher/spec path, preserved baseline policy, and thinnest executable implementation route.

picard1701ent Bundle

Enforce strict task, data, method, and evaluation boundary checks for paper or model reproduction before implementation or long runs. Use when reproducing papers, adapting methods to available data, classifying faithful versus adapted baselines, or requiring explicit user approval for any adaptive implementation.

mart337i Skill

Tell the agent to zoom out and give broader context or a higher-level perspective. Use when you're unfamiliar with a section of code or need to understand how it fits into the bigger picture.

mart337i Bundle

Grilling session that challenges your plan against the existing domain model, sharpens terminology, and updates documentation (CONTEXT.md, ADRs) inline as decisions crystallise. Use when the user wants to stress-test a plan against their project's language and documented decisions.

tvishwanadha Skill

Use Codex MCP for plan review, completion verification, and code review on multi-file changes or after context compactions.

sashakucharczyk Bundle

Surface and manage load-bearing assumptions before giving substantive analysis, recommendations, prioritization, forecasts, estimates, strategy, research synthesis, product/customer interpretation, or any answer where hidden assumptions could materially change the conclusion. Use when the user asks to align on approach, identify assumptions, make reasoning auditable, distinguish user inputs from inferences/model judgment/model knowledge, flag missing inputs, interpret ambiguous terminology, analyze customer signals, or stress-test assumptions.

sashakucharczyk Bundle

Use when Codex should delegate heavyweight reasoning, research, synthesis, drafting, critique, long-context analysis, or second-opinion work to ChatGPT WebApp through the Chrome plugin/skill while Codex remains the orchestrator. Trigger when the user asks to use ChatGPT WebApp, the best ChatGPT model, Agent, Deep Research, specialized ChatGPT modes, or to offload high-compute/high-token work from Codex to ChatGPT and have Codex evaluate, iterate, and integrate the result.

woosook0127 Skill

Use when a code, model, automation, or documentation change needs an explicit Codex-native implementation plan with assumptions, target files, acceptance criteria, and verification before execution.

mondlak1 Bundle

Consent-gated webcam access so the agent can see a single still frame from the user's camera while the user watches a live preview. NO VIDEO CAPTURE — every frame on disk is an explicit, on-demand PNG. Use for hardware testing, OCR, plant logs, posture coaching, instrument reading. Includes `adjust` for tweaking camera properties (brightness, exposure, focus) live.

hmzainjamil Skill

G0DM0D3 — Ultra-Model Racing (55 Models)

hmzainjamil Bundle

Orchestrates all 210 specialist agents simultaneously. Auto-discovers skills, MCP servers, repos. Used for comprehensive/360/full-analysis tasks.

laxmimerit Skill

Use this skill whenever building, assembling, or creating any 3D model in Blender via MCP tools (mcp__blender__execute_blender_code). Covers all object types: furniture, vehicles, architecture, characters, props, mechanical parts, scenes. Trigger any time Blender Python code will create geometry with primitives, bmesh, or modifiers. Even simple objects benefit from correct primitive math and verification. Always invoke this skill before writing any Blender geometry code - it prevents the three most common geometry bugs that cause exploded-looking models.

laxmimerit Skill

Use this skill when generating a 3D model from an image or text prompt via Hunyuan3D MCP tools and importing the result into Blender. Trigger whenever mcp__hunyuan3d tools are involved OR when the user says 'generate from image', 'generate from text', 'AI 3D model', or 'Hunyuan3D'.

laxmimerit Skill

Use this skill whenever generating, reconstructing, analyzing, or building a 3D character from one or more reference images inside Blender using MCP tools. Trigger this skill before ANY Blender character generation workflow — even if the user just says 'build this character', 'make this in 3D', 'create this model', or pastes a reference image. Covers stylized, anime, realistic, game-ready, cartoon, low-poly, high-poly, humanoid, creature, mascot, and cinematic characters from any source: screenshots, concept art, AI-generated images, sketches, or photos. This skill ensures correct proportion extraction, topology planning, staged construction, silhouette preservation, and animation-safe geometry. Do not skip this skill even for 'simple' characters.

hmzainjamil Skill

Website Builder — $10K Site From One Prompt

hmzainjamil Skill

Prompt Shortcuts — Slash Command Library

hmzainjamil Bundle

Ultra-fast token-efficient execution. token-turbo + caveman + parallel blast + multi-model routing. Active by default — no command needed.

3taste Bundle

单机双 Agent 开发审查工作流：开发者 Agent 逐模块实现代码并提交， 审查者 Agent 异步轮询每次提交、输出审查结论， 基于信号文件驱动完整的开发→审查→合并循环。

4karam Bundle

Concise blueprint for building MCP servers with FastMCP (Python). Covers core primitives (tools, resources, prompts), Context injection, lifespan management, server composition, transports, deployment, and security. Use when creating, reviewing, or deploying any MCP server.

joshbricel Bundle

Migrate a Tableau workbook (.twbx) to Omni: unpack the file, mine the XML for every calc / parameter / dashboard zone / action / style, then rebuild as an Omni dashboard via the official `omni` CLI. Outputs a structured extraction bundle, an Omni import payload, and a parity report. TRIGGER when: "migrate this Tableau dashboard to Omni", "convert this .twbx", user provides a .twbx path, /tableau-to-omni invoked. SKIP for: building net-new Omni dashboards from scratch (use omni-vega-chart); branching the model alone (use omni-branch-creator); Looker / Power BI conversions (out of scope).