AI & ML

sxvan Skill

Teach the user the domain behind a feature before any design or code exists. Run it as the first step of a feature, before architecture, planning, or the first implementation prompt. Two phases in one command; a short assessment to find where to start, then one teaching message. Domain knowledge only, never the codebase.

sxvan Bundle

Upgrade Odoo modules from version 18.0 to version 19.0. Use it when a module, manifest, model, view, controller, or test written for Odoo 18 has to run on Odoo 19, or when the user asks what changed in the ORM between the two versions.

catlog22 Bundle

Universal team coordination skill with dynamic role generation. Uses team-worker agent architecture with role-spec files. Only coordinator is built-in -- all worker roles are generated at runtime as role-specs and spawned via team-worker agent. Beat/cadence model for orchestration. Triggers on "Team Coordinate ".

catlog22 Bundle

Full lifecycle team skill — plan, develop, test, review in one coordinated session. Role-based architecture with coordinator-driven beat model. Triggers on "team lifecycle v4".

dmoreq Skill

Fetch version-sensitive library/API docs via Context7 MCP (setup, examples, method usage). Not for general CS concepts or SOTA surveys.

makinotes Bundle

EdgeLab · 用可审计公式计算 EV、Kelly 仓位、策略结构数学（bull put spread 等）。触发：/elab-model、「算EV」「期望收益」「Kelly」「策略数学」「这单结构怎么样」「收得薄不薄」 EdgeLab · Auditable-formula model layer: EV, Kelly sizing, and options strategy math. Trigger: /elab-model, "calculate EV", "expected value", "Kelly", "strategy math", "how does this structure look"

makinotes Bundle

EdgeLab 投研编排器。按问题选择财报、产业链、估值、事件、波动率、拥挤度或宏观研究；可按用户提供的方法执行，或在明确接口与权限后委派外部研究 Agent。支持按指定连接读取富途、长桥、IBKR 数据，综合带证据与反证的中性研报。不给买卖方向。 触发方式：/elab-research、/投研、「深度研究一下 X」「用富途/长桥/IBKR 数据研究 X」「研究 X 的期权面」「拉个多空辩论」 EdgeLab research orchestrator: pluggable tools + self-refutation + optional real multi-agent debate (bull/bear + independent judge). Trigger: /elab-research, "deep-research this ticker", "run a bull/bear debate"

ar9av Bundle

The foundational knowledge distillation pattern for building and maintaining an AI-powered Obsidian wiki. Based on Andrej Karpathy's LLM Wiki architecture. Use this skill whenever the user wants to understand the wiki pattern, set up a new knowledge base, or needs guidance on the three-layer architecture (raw sources → wiki → schema). Also use when discussing knowledge management strategy, wiki structure decisions, or how to organize distilled knowledge. This is the "theory" skill — other skills handle specific operations (ingesting, querying, linting).

xinbenlv Bundle

Rules and a mandatory pre-send check that make an agent's natural-language output readable to a human reader. Load this skill before you write any prose a human will read - a status line, a progress report, an explanation, a plan, a summary, a chat reply. Also load it when the user says your writing is hard to understand, vague, telegraphic, clipped, full of jargon, metaphors or invented words, or when the user says 看不懂, 说人话, 你的中文很难懂, 把话说完整, 不知道你在说什么. Cross-language rules live in this file. Load lang/zh/PRINCIPLES.md for Chinese output and lang/en/PRINCIPLES.md for English output.

svnscha Skill

Triage a Windows crash dump - identify the exception, the faulting frame, and what to look at next. Use when the user points at a .dmp file or asks why a Windows process crashed.

svnscha Skill

Attach to a running Windows process through a WinDbg debug server and inspect it live. Use when the user wants to debug a process on another machine, or one already under a .server session, rather than a crash dump.

svnscha Skill

Attach to a live Windows kernel target over KDNET, a named pipe, or serial and drive it. Use when the user wants to debug a kernel, a driver, or a bugchecking VM.

svnscha Skill

Check that this machine can actually debug - MCP connection reachable, debugger present, symbols configured - and explain how to fix whatever is missing. Use when mcp-windbg tools fail, when a session will not open, or before a first debugging session.

denshoapp Skill

Read, write and organise pages in a Densho workspace through its MCP server. Use when the user asks to look something up in their wiki or documentation, draft or update a page, answer from what the team has written down, leave feedback on a page, or tidy the wiki (labels, icons, review status, moving, duplicating, trashing and restoring pages, page history, public links, subscriptions). Covers pages, spaces, comments and the Densho Markdown dialect.

hkuds Bundle

Inspect and optionally adjust the agent's runtime state. Use to check the current model or preset, context window and runtime limits, workspace and tool configuration, subagent status, and request routing metadata such as channel, chat ID, and sender ID; diagnose unavailable capabilities; change allowed runtime settings; or store temporary session scratchpad values.

akashsoti Bundle

Guides the user step by step to create a UX outcome statement and its outcome-driven metrics (success, progress, problem-value), starting from the business outcome and refusing to skip gates. Use this skill whenever the user wants to write, draft, refine, or review an outcome statement, UX outcome, success metric, progress metric, problem-value metric, baseline, threshold, party moment, or pre-registration for a feature or project — or says things like "let's define the outcome for X", "help me create metrics for this feature", "what should the success metric be", "draft the outcome statement", "set the baseline", or names a Pine One project (settlements, VAS, ODS/SDS, BACR, onboarding, hardware diagnostics, Ads Agent) together with goals, metrics, or measurement. Also use it when the user pastes a business goal or PM brief and asks how design should measure its contribution.

hypit-ai Bundle

Give your Agent a reference video or a brief and receive an editable video project.

bighadj22 Skill

Update runbook for a self-hosted CodFlow install — an AI agent following it fetches the latest code from the CodFlow GitHub repo, merges it into an EXISTING checkout, syncs the gitignored wrangler.toml / .env / .dev.vars files with any template changes (keeping the user's resource IDs, worker names, and domains), creates ONLY the new Cloudflare resources an update introduces, applies new D1 migrations without touching data, and rebuilds + redeploys each updated worker (server, dashboard, or storefront theme) — never re-running the full setup, never regenerating secrets, never re-seeding a live store, and reporting every step to the user in plain language (they may be a non-developer). Use when the developer already set CodFlow up (via the codflow-setup skill) and wants to update or upgrade to the latest version from https://github.com/bighadj22/codflow (branch main), pull the latest changes/release, apply new migrations after pulling, or asks how to get updates for their existing install.

hypit-ai Bundle

把参考视频或创作 Brief 交给 Agent，得到可继续编辑的视频作品。

dmoreq Skill

Implement or modernize Python modules with typing, uv, ruff, pyright, and pyproject.toml. Not for framework choice, concurrency model, runtime profiling, or PR review.

jinneverett Skill

IELTS Writing grading coach. Four-criteria scoring + sentence-level annotation + rewrite comparison + prompt analysis + history tracking. Triggers: /ielts-writing, "grade my essay", "take a look at this", "analyze this prompt", "writing practice"

hermes-brasil Skill

Operar Hermes Agent em produção no VPS — update, restart de gateway, health-check, profiles múltiplos e ambientes sem systemd user bus.

hermes-brasil Skill

Criar assistente RAG com conhecimento da empresa.

firstp1ck Bundle

Bound multi-step agent work with one outcome, explicit resources, short horizons, stop conditions, and escalation instead of looping or silently expanding permissions. Do not use for one read, one deterministic edit, simple explanations, or multi-agent governance.

firstp1ck Bundle

Plan and report a bounded comparison of fixed small-model task fixtures without claiming unrun live results or substituting a model/provider. Do not use for normal task execution.

kozaksv Bundle

Read and maintain a project's LLM Wiki: query, init, ingest, edit, lint, cleanup, split, protect and status. Use for wiki/вікі requests and project-specific architecture, setup, decisions and recipes. Read relevant wiki evidence before answering; use the GitHub adapter for remote repos.

agentworkforce Skill

Use when writing agent-relay workflows that must fully validate features end-to-end before merging. Covers the 80-to-100 pattern - going beyond "code compiles" to "feature works, tested E2E locally." Includes repair-before-failure validation gates, mandatory sequential Claude-then-Codex fresh-eyes review/fix loops with test hardening, PGlite for in-memory Postgres testing, mock sandbox patterns, test-fix-rerun loops, verify gates after every edit, and the full lifecycle from implementation through passing tests to commit.

aossie-org Bundle

Guides developer agents on how to leverage and initialize Model Context Protocol (MCP) servers (GitHub & Puppeteer) to automate code changes, UI/UX testing, and PR management in AOSSIE projects.

cognigy Bundle

Add a new MCP tool (or extend an existing one) in the NiCE Cognigy Plugin's MCP server

cognigy Skill

Use when creating or configuring Cognigy agent tools — choosing the tool type (tool, http, mcp, knowledge, send_email) and their configuration schemas.

career-cafe Skill

Enforces strictly emoji-free, concise, objective, and technically rigorous communication standards across all AI agent interactions, prohibiting decorative emojis, conversational filler, and informal preambles.

altryne Bundle

Use Jev as System One to save the user reasoning-model tokens, time, and cost on bulk reading or repeated semantic decisions. Trigger when sifting large documents or tool outputs, selecting evidence or actions from many candidates, or checking many items against criteria, even if the user never mentions Jev. Also use for "Jevify this", Jev question design, and product work where frequent, inexpensive judgments could improve the experience.

job-yang Bundle

面向验证的开发（VDD）的完整方法，也是 VDD 理论的最小可复现证明——装上它，任何一个光秃秃的 Agent 都能立刻跑出"不乱报喜、不硬挤问题、敢说没问题"的效果，亲身验到 VDD 真的有用。它做的是：在你说"我懂了""做完了""搞定了""没问题"之前，先逼自己走一遍验证，把爱报喜、没做好也说做好了的毛病摁住；同时在多轮、大改动里守住一把不随对话漂移的尺子，既不放过真问题，也不为了显得勤奋硬挤问题、过度修改。任何一次改代码、查 bug、给结论、下判断、回答一个你没十足把握的问题时都该用，尤其是你正准备说"完成了""应该没问题了""我明白你的意思了"、或者被要求"整体 review 一下""从全局看看有没有问题"的时候。也适用于用户明确要求"验一下""你确定吗""VDD 走一遍""自证一下""别急着说完成"。别只在被点名时才用；只要你即将声称一件事做成了、或即将采信一个前提、或即将对一大片改动下"有没有问题"的判断，就主动走这套流程。这套方法完全靠你自己就能执行，不依赖任何外部工具或环境。

cognigy Skill

Use when the user wants to build, create, or set up a new Cognigy AI Agent from scratch — covers listing projects, ensuring an LLM exists, creating the agent, testing it, and refining persona/job fields.

cognigy Bundle

Use when the user wants to red-team, adversarially test, jailbreak-test, or probe the guardrails of a Cognigy AI Agent — covers scoping consent, deriving the agent's policy surface from its config, generating targeted probes, scoring verdicts, and producing a standardised findings report.

cognigy Skill

Use when the user wants to add knowledge, RAG, or a knowledge store/source to a Cognigy agent — covers embedding vs Knowledge Search models, Knowledge AI settings, ingesting sources, and attaching knowledge as a tool.