AI & ML

counterfactual5 Skill

Standing CSP local-environment handbook for Claude Science Proxy. Covers (1) web access — no hosted Web Search; ALWAYS use the local `web-search` MCP (GENERAL lane `csp_web_search`, LITERATURE lane `search_literature`, then `fetch_url`) and NEVER the hosted Anthropic `web_search` tool; (2) Science `search_skills` must always pass non-empty `query` or `prefix`; (3) filesystem — never `/mnt/data`, save to workspace cwd then `save_artifacts([...])`; (4) CJK matplotlib fonts; (5) skills/env — don't rely on `host.skills.publish()`, use analysis `python` for science packages; (6) network allowlist via CSP Start + `~/.csp/network-allowlist.json`.

cryptoyasenka Bundle

Drill the canonical FHEVM async-decryption flow — flag with `makePubliclyDecryptable`, accept a KMS-signed cleartext callback, verify with `checkSignatures` BEFORE writing state. Use when an AI agent is asked to (a) write a Solidity contract that decrypts encrypted state via the KMS, (b) write or audit a callback that takes `(bytes abiEncodedCleartexts, bytes decryptionProof)`, (c) review an FHEVM contract for replay or fake-decryption bugs, (d) explain why mainnet has no synchronous decrypt.

cursor-demo-lab Skill

Use subagents. Runs TDD with two subagents: test-writer then implementer. Requires an IDE Composer chat so Task goes through TaskToolCallHandler and the Atlassian unit-test hooks fire. Skip in CLI / agent-exec. Skip when the test path is unclear, expensive, or not requested.

cyog941 Bundle

Build and test with the gauntlet-loop technique. Convert a substantial prompt into a bounded worker/evaluator fan-out, then set it up and execute it end-to-end at the highest quality the session budget allows, tiering models per role (top-tier director/verifier where judgment decides the outcome; cheaper workers elsewhere). Use for requests mentioning a gauntlet/guantlet loop, worker-critic loops, fan-out subagents, multi-agent orchestration, parallel workers, iterative refinement, or building a complex deliverable through delegated work with evaluators and revision loops. Produce a prompt-only conversion when the user explicitly asks for just the prompt or spec.

d-kimuson Skill

Run the gatecheck release flow end-to-end. Use when the user asks to release patch, minor, major, beta, or an explicit semver; includes non-interactive release execution, GitHub Actions monitoring with gh, draft release note cleanup, and publishing the GitHub Release.

daddongoba Bundle

Unreal Engine MCP (Model Context Protocol) development skill for UE5.8+. Covers: editor control (spawn/remove actors, scene & level ops), blueprint authoring via DSL (components, variables, events, graphs), collision/trigger/physics interactions, materials, UMG, save systems, keyframed animation & ControlRig posing via Sequencer toolsets (~290 tools incl. FBX in/out), Niagara VFX, PCG graphs, physics assets, state trees & behavior trees. Also: programmatic batch orchestration, PIE control & log forensics, screen automation fallback, verified pitfalls/templates/design-conventions knowledge base. Use for ANY UE editor task driven over the MCP server (port 8000).

darasoba Skill

Model relay

darfaz Bundle

Real-time AI agent security scanner. Detects prompt injection, jailbreak attempts, credential/secret leaks, PII exposure, and dangerous tool calls. Activate when: (1) scanning inbound messages or tool outputs for prompt injection, (2) checking outbound content for credential leaks or PII, (3) auditing agent session logs for security events, (4) evaluating tool call safety before execution, (5) user asks about security scanning or threat detection. Covers OWASP Top 10 Agentic AI risks.

darkamenosa Bundle

Generate or edit raster images by calling the ChatGPT/Codex hosted image_generation flow with local Codex or OpenClaw OAuth credentials, then save decoded image files for OpenClaw and other agent workflows.

davidho27941 Skill

Practical usage notes and known pitfalls for the mcp-tradingview MCP server (rate limits, alert activation, data quality caveats). Use whenever calling any mcp__mcp-tradingview__* tool — querying quotes, financials, technicals, forecasts, news, screeners, or managing alerts and watchlists, especially for Taiwan (TWSE/TPEX) symbols.

davistroy Skill

Use when checking on Jetson Orin Nano inference performance landscape — scans llama.cpp releases, JetPack updates, Qwen models, NVIDIA Jetson forum, and community GGUF builds for actionable changes. Run periodically from the jetson project directory.

ddgod123 Bundle

北京高端住宅租赁顾问（RentPro）。当用户想租房、找房、询问某个小区/商圈/区域有没有在租房源、想比较几套候选房源，或想联系房源经理核验与约看时使用。常见说法包括：租房、找房、在租、月租、预算、几居、面积、平层、大平层、别墅、服务式公寓、公寓、国贸、望京、朝阳、海淀、东城、西城、中央别墅区、通勤、地铁、学校、泳池、会所、层高、宠物、精装。不处理买房、卖房、家装、贷款等非租赁需求，也不做楼盘行情分析；所有房源事实只通过 RentPro MCP 获取，不读取数据库、不编造。

folajj Bundle

Use when converting business workflows, API docs, or MCP handoff materials into an orchestration-heavy agent skill focused on tool routing, branching, loops, retries, pagination, and exit handling, or when reviewing/scoring that kind of orchestration-heavy skill. Triggers include 业务流程转skill、接口文档转MCP交接、工具路由设计、编排型技能评审、MCP编排评分.

grubbylee Bundle

Use when users ask which already-installed local Agent Skill should handle a task. It calls local skm recommendations from the real installed skill catalog; it does not perform the task itself. 当用户询问“做某件事应该用哪款已安装 skill”时使用。它通过本机 skm 基于真实已安装 skill 清单做推荐，本身不执行具体任务。

hruiccc Bundle

Generate interactive 3D holographic collectible-card websites from a user description or reference image, using layered artwork, Blender and Three.js. Includes project-local Blender installation, reusable parallax materials and browser verification. Host-agnostic — any harness that runs a multimodal model can use it, since the model draws the layer artwork and inspects the rendered frames. Use when the user asks for 全息闪卡, 镭射卡, 3D 卡牌网页, a holographic collectible card site, or an editable card.blend.

inference1 Bundle

Grounded in first principles, rigorously examine and refine a user's plan, task, decision, goal, strategy, proposal, or idea through structured, progressively deeper questioning, in order to bridge the gap between the User and the Agent. Use when the user explicitly requests grilling, challenge, pressure-testing, cross-examination, red-team review, pre-mortem analysis, or a decision audit. The goal is to uncover unclear objectives, hidden assumptions, contradictions, weak evidence, missing information, overlooked constraints, dependencies, risks, trade-offs, failure modes, and misalignment between intended outcomes and likely real-world results. Begin by establishing a shared understanding of the user's actual intent, goals, constraints, and success criteria. Ask focused, high-leverage questions rather than broad or repetitive ones. Adapt each question based on previous answers, probing deeper where uncertainty, unsupported assumptions, or strategic weaknesses remain. Regularly summarize the current understan

jokerjohn Skill

Use when XiaoHongShu notes appear published in note manager but cannot be opened by direct link or seen by other accounts, or when xhs-agent content must be rewritten to avoid finance/trading visibility risk.

keep-maker Bundle

Agent Token 效率综合指南：整合 Ponytail/Caveman/Headroom 三件套策略，以及输入压缩、输出精简、代码最小化、上下文管理、工具调用、记忆与工程规范等省 token 方法。在用户问如何省 token、降低 API 成本、优化 Agent 上下文、或编写/审查高 token 消耗工作流时使用。触发词：省token、token优化、省钱、Headroom、Caveman、Ponytail、上下文压缩、API成本。

langevc Bundle

Drei-Stufen-Humanizer für deutschsprachige Texte. Entfernt KI-typische Muster und macht Texte natürlicher, menschlicher und kontextgerecht lesbar. Basiert auf der Wikipedia KI-Erkennung v1.35 – adaptiert für allgemeine deutschsprachige Texte. Stufen: 1) ANALYSE – evidenzbasierte KI-Muster-Erkennung nach Wikipedia v1.35 Kriterien 2) RECOMMEND – konkrete, belegbasierte Änderungsempfehlungen pro Kriterium 3) FINETUNE – zielgerichtete Anwendung mit konfigurierbaren Stil-Reglern Zukünftig erweiterbar auf die Top X weltweit meistgesprochenen Sprachen.

learnprompt Bundle

Verified prompt templates and structures for Seedance 2.5 / 2.0 video generation, distilled from the highest-performing published cases on goodcase.ai. Use when the user asks to write, revise, or review a Seedance video prompt, or asks how to prompt Seedance for a specific kind of shot (narrative short, ad/commercial, UGC talk-to-camera, dialogue with lip-sync, action, product reveal, and similar). Also use when the user names Seedance, 即梦视频, or Seedance 2.5/2.0 directly.

mikehutu Skill

Drive MiniMax Music3 (text-to-music) and MiniMax H3 (text/image-to-video) through a local/remote ComfyUI on DGX-Spark hardware. Use when generating music, songs, beats, instrumentals, or AI video from prompt/lyrics; or when converting them to MP3/MP4. Portable + ENV-driven, no personal config.

mimirs402 Bundle

Use for: why did a GitHub Actions run fail (root cause); retry a failed workflow; is a GitHub bounty open; which bounty to work on; audit AGENTS.md; will MCP tools/list changes break clients.

mirannonarbitrable290 Bundle

AI Agent-driven Kaggle competition workflow. Learn from real competition experience: score stabilization patterns, submission troubleshooting, kernel workflows, GPU task delegation, and the spec-driven development approach that achieved top leaderboard positions. Use when: working on any Kaggle competition, analyzing submission failures, setting up automated pipelines, or replicating top notebook solutions.

momen-walied Bundle

Design, implement, review, debug, optimize, migrate, and operate production RAG systems. Use for repository work involving ingestion, parsing, chunking, embeddings, indexes, retrieval, reranking, grounded generation, citations, evaluation, tracing, monitoring, latency, cost, fallbacks, ACLs, multi-tenancy, large data, incidents, and vendor migrations. Analyze existing implementations top-down, derive failure scenarios, make verified code changes, and continue from plans into implementation when requested. Compose with repository instructions, Spec Kit, OpenSpec, other skills, MCP tools, hooks, and native agent orchestration instead of replacing them. Keep small changes small and high-risk changes evidence-gated. Do not use for prompting or fine-tuning tasks without external retrieval.

moonweave-systems Bundle

User-invoked mid-session audit skill. Detects (1) drift between current task and original session intent via transcript analysis, and (2) codebase rot — git uncommitted accumulation, test/lint state from cached artifacts, file bloat, TODO accumulation, circular dependencies, module boundary degradation. Manually triggered via `/compass` only — never auto-invoked. Use when the user types `/compass` optionally followed by a baseline intent string. Without arguments, derive baseline via cascade — most recent spec (24h), then task-init heuristic, then explicit user prompt. Never use the literal first user message of the transcript as baseline.

mr-funny Bundle

Search, compare, and download stock images from Pexels, Unsplash, and Pixabay, stock videos from Pexels and Pixabay, and browser-selected Pixabay music through the hbg-free-material CLI. Use when an agent needs licensed stock media candidates, preview URLs, creator attribution, source links, local files, quality selection, or reproducible manifest records without loading an MCP server.

nimachu Bundle

Create complete Xiaohongshu image-and-text posts from a topic, article, document, or rough idea, including research, fact checking, post copy, pagination, storyboards, a 3:4 cover, 9:16 content pages, image generation or fully local SVG-to-PNG fallback, and final quality review. Use for 小红书图文、小红书封面、知识卡片、漫画科普、文章转图片、AI/科技科普图组、逐页生图提示词, or revisions to an existing Xiaohongshu series. Prefer an available image-generation model; use the bundled local renderer only when no image-generation tool is available. Apply an SVG text patch only after the user explicitly reports incorrect or unreadable text in an image.

nomosgrowth Bundle

Query the OpenNomos MCP REST layer with the default base URL `https://api.opennomos.com` and an `nk_` API key. Use when tasks involve analyzing OpenNomos projects, available tasks, project overview, daily operational metrics, project Opportunity reports, project-side Opportunity review updates, current actor identity, my submitted Opportunity feedback, my task event stream, user points, or points ledger from the live backend.

orbbitai Bundle

Connect to and query Orbbit's Aggre MCP server — the org's data platform, holding Taiwan and China law and case law, US FDA drug/device data, SEC filings, congressional and insider trading, and market data. Use this whenever the user wants to search or browse Aggre's ingested data, connect an MCP client (Claude Desktop, Claude Code, etc.) to Aggre, build or debug the aggre-mcp stdio server, or call Aggre's MCP tools (list_datasets, search_dataset, retrieve, get_record, get_dataset_schema, export_sample, export_dataset, get_politician_trades, get_investor_holdings, get_insider_trades, who_traded) directly via REST. Also trigger for questions like "how do I query our legal data from Claude", "what does the FDA say about this drug", "which senators traded NVDA", "connect MCP to aggre.orbbit.ai", or "what MCP tools does Aggre expose."

perrylink Bundle

Use when developing, reviewing, packaging, debugging, or answering questions about DeepSeek Harness (DSH) plugins — the plugin-based agent harness on vendored Cordis. Applies the official plugin-development constraints (plugin contract, cordis.yml layers, services/events/effects, tool DSL, bundles/profiles) backed by the dsh-plugin-guide knowledge base.

princeofchum Bundle

Expert guidance on Zoho MCP — Zoho's service for building Model Context Protocol (MCP) servers that expose 300+ Zoho and third-party apps as tools to AI clients like Claude, ChatGPT, Cursor, Windsurf, and VS Code. Use this skill whenever the user mentions Zoho MCP, the Zoho MCP Console, creating/configuring an MCP server on Zoho, adding Zoho tools (CRM, Mail, etc.) to an AI agent, connecting Zoho to Claude/ChatGPT/Cursor, Zoho MCP authorization (on-demand vs. connections), the MCP URL/API key, collaborators, logs, or troubleshooting why a Zoho MCP server won't connect or a tool won't run — even if they don't say the word "MCP" explicitly. Pay special attention to the platform's limitations, which are covered in detail here.

purapura-pp Skill

指导 AI 编码助手以资深工程师的方式理解任务、调查上下文、规划、执行、验证和汇报。规定指令优先级（用户最新消息高于系统注入的"执行计划"指令）、todo 必须带完成标准与证据指针、汇报断言必须带 [实测]/[仿真]/[推断]/[未验] 标签、修 bug 先复现、能写成断言的验证不靠肉眼看、阶段心跳与提交保存点、开工环境快照。含处理用户多轮反馈（尤其附截图、照片、日志）的方法，以及 UI/视觉、配置/构建、真机/硬件三类工作的专项验证清单。适用于任何读写代码、操作文件、运行命令或回答技术问题的任务，尤其在开始多步骤工作、执行已批准的计划、或处理用户对上一轮成果的反馈之前。

pushary Skill

Push notifications and human-in-the-loop for AI agents. Use this whenever a running agent needs a human and nobody is at the terminal, such as before an irreversible or destructive action, before spending money, deploying, force-pushing or deleting, when blocked on a decision outside your authority, when running unattended and you hit a genuine ambiguity, when another skill's workflow says to confirm with the user, and when a long task finishes or fails with nobody watching. Also use it when the user says things like keep going and ping me on my phone if you need anything, notify me when my agent needs me, approve from my phone, ask me questions while I am away from the terminal, run this overnight, keep working while I am in a meeting, I am stepping away, do not wait for me, or wants a long task to run unattended. Send alerts when tasks finish or fail, ask questions (yes/no, multiple choice, or free text) via push, and get answers from the user's lock screen. Use these tools proactively - do not wait for the

rty798 Bundle

Structured long-form fiction workflow with chapter-type routing, factual state, and narrative-first review.

rexysaragih Bundle

Build enterprise-ready TypeScript (default) or Python MCP servers that wrap real APIs for AI agents. Researches API/SDK options, defaults to read-only v1, requires security annotations/tests/timeouts, scopes a small tool surface, designs md/json outputs, scaffolds clients/tools like mcp-bitbucket and mcp-confluence, then runs mandatory audit/QC (mcp-maker qc) before shipping. Use when the user asks to create, design, scaffold, implement, audit, or QC an MCP server, MCP tools, Model Context Protocol integration, or wrap a SaaS/API for agents.

ryan-7962 Bundle

Use this skill for pregnancy Q&A, local pregnancy memory, current medical state, medically relevant red/yellow/green triage, Xiaohongshu content audits, baby weekly diary, message-channel adapter work, and upgrade-safe pregnancy data handling. Optional extensions include partner summaries and dad diary. Trigger whenever the user asks to build, run, customize, or use a Pregnancy Copilot style agent skill.