AI & ML

tomevault-io Bundle

Instructions for updating the README.md file in the `dart_mcp_server` package. Use when this capability is needed.

tomevault-io Bundle

Run PinchBench benchmarks to evaluate OpenClaw agent performance across real-world tasks. Use when testing model capabilities, comparing models, submitting benchmark results to the leaderboard, or checking how well your OpenClaw setup handles calendar, email, research, coding, and multi-step workflows. Use when this capability is needed.

tomevault-io Bundle

Debug and fix failing tests using Wallaby.js MCP tools. Use when investigating test failures, debugging test errors, inspecting runtime values, checking code coverage, or updating snapshots. Use when this capability is needed.

tomevault-io Bundle

Local Zod v3 documentation reference. Use when asked about Zod schema validation, TypeScript type inference, error handling, or migration from earlier versions. Use when this capability is needed.

88plug Skill

See and operate the user's real, physical Linux desktop through the mcp-screen MCP — screenshot any monitor and click/type/scroll/drag any visible app. Prefer screen_watch (~1 fps) after graphs/maps/canvases/loaders so thrashing UIs are caught like a human watches, not one freeze-frame. Use whenever the user points at something on their monitor(s) or wants you to read/drive a real GUI (Slack, browser, terminal, dashboards). Prefer over app/web APIs when looking at the real screen. NOT for pasted images, local files, shell-only, or remote SSH/TUI. Encodes locate → ground → act → confirm/watch plus hard-won gotchas.

8sign24 Bundle

为 Flink 项目（CDC 同步作业 / Flink 流计算作业，多作业单仓库或多仓库）生成或维护 `.agent/` 目录——多 Agent 共享的渐进式披露项目上下文：每个作业一个目录按生命周期分文件（README 接入/plan 规划/dev 开发/test 测试/acceptance 验收/changelog 迭代），加全局代码目录规范、任务看板、决策/调试/交接知识库、公共工具沉淀。Use when the user asks to create, scaffold, initial or update a `.agent` directory (or .agent 目录, agent 协作目录, 多 agent 上下文, progressive disclosure docs) for a Flink project, or asks to organize Flink project conventions for multiple coding agents (Codex/Claude/Cline/Cursor), shared task context, code layout conventions, or common utility settlement, or wants to remove tool-specific rules files (like CLAUDE.md) and consolidate everything into the .agent directory.

aiops-tools Bundle

Use this skill whenever the user needs to observe or govern on-endpoint local LLMs running on Ollama, llama.cpp (llama-server), LM Studio, or a local single-node vLLM — inventory installed/running models with an allow/deny verdict (shadow-AI detection), inspect VRAM residency, model license/params/capabilities and server version, view the model policy, detect model provenance/digest drift (re-pulled or tampered weights; strong for Ollama/llama.cpp, id-only and honestly weaker for LM Studio/vLLM), scan a prompt for secrets / PII / source-code / jailbreak with a weighted risk band, route a prompt THROUGH a guard that scans + policy-gates + records + runs-if-allowed (guarded_generate / observe_chat), query the observed-usage log, and roll up anomalies (shadow models, digest drift, high-risk + blocked prompts). Always use this skill for "what local models are installed", "find shadow / unsanctioned AI models", "which model is loaded in VRAM", "scan this prompt for secrets/PII before sending", "stop secrets leakin

aiops-tools Bundle

Use this skill whenever the user needs to manage VMs and containers on Proxmox VE — list/inspect/configure VMs, power and lifecycle (start/stop/shutdown/reboot/reconfigure/clone/delete/migrate), snapshots (create/delete/list/rollback), disk grow/move, vzdump backups (create/list/restore), LXC containers (list/start/stop), cluster/node status, cluster resource inventory, async task polling + logs, free-VMID lookup, HA status, resource pools, firewall inspection, guest-agent ping, and storage listing. Also use it to diagnose cluster health — rank nodes by CPU/memory/disk pressure and scan guests for saturation (read-only RCA). Always use this skill for "list proxmox vms", "start proxmox vm", "stop proxmox vm", "proxmox snapshot", "proxmox backup", "restore proxmox vm", "resize proxmox disk", "proxmox vm status", "migrate proxmox vm", "proxmox container", "proxmox ha", "proxmox pool", "proxmox firewall", "list proxmox storage", "proxmox node pressure", or "why is proxmox slow" when the context is explicitly Prox

abolfazlshahi Bundle

Procedural operating layer for custom-provider and third-party models working in Claude Desktop, Cowork, and Claude Code-style Agent Skills runtimes. Teach the model to identify its runtime first, discover and probe capabilities, remediate missing integrations when possible, route tools, operate Browser/Chrome/Computer Use, use MCP/connectors/files/Git/code, respect provider and execution boundaries, verify outcomes, and recover safely. Use for coding, web apps, browser automation, GUI work, research, integration tasks, and complex multi-step execution when correct runtime and tool selection matter.

adityapal67 Bundle

Compress any conversation into a portable JSON context graph for handoff to a new chat or LLM. Trigger on: "compress this chat", "save context", "new chat", "hand off", "transfer context", "too many tokens", "start fresh", "export conversation", "context graph", "continue in new session". Also trigger proactively when the user mentions starting over in a long session.

adrianadem Bundle

Prompt Engineer

advent-ai Bundle

Discover and call TrueNorth agent tools dynamically through the tn CLI.

aethon-chrysean Bundle

Rigorous problem-solving via bilingual research (Chinese + English), proactive query expansion, and interdisciplinary reasoning. Challenges before confirming, connects distant domains, never fabricates. Triggers on: any request for a solution, explanation, analysis, recommendation, comparison, debugging, architectural decision, or concept clarification. Also triggers on: "how does X work", "why does X happen", "best approach for", "help me understand", "explain X", "compare X and Y", "I'm stuck on", "what am I missing", any problem-solving prompt, or any question where accuracy matters. Triggers on casual phrasing like "what's the deal with X". Does NOT trigger for pure creative writing, simple greetings, or tasks handled by more specific skills (st-formula-explainer, st-framework-to-pytorch, spatial-transcriptomics-validator).

agentgamelab Bundle

面向 Agent workspace（OpenClaw / Claude Code / LangGraph / 任何有"系统配置文件 + 运行日志"概念的框架）的自进化 skill——一次运行走完整闭环：追踪前沿研究 → 精读分析 → 系统自检 → 按 L1/L2/L3 风险分级自主执行或提案。触发词："运行自进化"、"跑一次 harness evolve"、"检查一下系统有没有能优化的地方"、"今天有什么新研究可以落地"、"帮我看看 agent 配置是不是该收敛了"。

agentgamelab Bundle

TuneUp — 让 AI 真正理解你说的每一句话。把模糊需求翻译成高信息密度 Prompt 的工具，含智能分流（快车道 / 主车道 / 急诊车道）+ 四象限锻造法（口语 / 模板 / XML / Lisp）+ 专业视角预设（domain expert lens 装备法）+ 理论锚点词典（理论名当高信号 token + 黑话翻译 + 误用理论黑名单核伪）+ 高风险领域安全边界。 触发场景分四层： （1）显式触发：用户说"帮我写个 prompt""优化指令""写 system prompt""Lisp 风格""结构化 prompt"等明确与 prompt 相关的话； （2）隐式触发：用户对 AI 输出不满但不知道问题出在 prompt 上——典型信号包括：用户在同一话题上连续追问、纠正、补充两轮以上，用户说"不是这个意思""我要的不是这个""你没理解""更具体一点""别给我废话""认真点"，用户把同一个需求换了措辞重新说了一遍，用户的追问越来越长越来越具体（说明 AI 一直没抓住重点）； （3）专业视角触发：用户做评估 / 诊断 / 创作 / 审查 / 挑刺 / 专业 brainstorming 类需要领域专家视角的任务 → 触发专业视角预设；用户做财务 / 法律 / 医疗 / 生死 / 不可逆决策类任务 → 硬触发专业视角预设 + 强制安全边界； （4）配方触发：用户想把反复使用的指令打磨成可复用模板，说"保存这个""做成模板""我以后还要用"。 核心原则：最需要 TuneUp 的用户往往不知道"prompt"这个词，他们只是在反复尝试让 AI 理解自己——识别到这种模式时必须主动介入。

ahaaad3759 Bundle

Prompt Engineering - 通用提示词工程助手

akash97p Bundle

Send local desktop notifications through AgentNotify when a coding agent needs human input, permission, a decision, or attention, or when it completes, fails, or becomes blocked. Use during autonomous coding work to surface actionable state changes from Windows, PowerShell, Command Prompt, WSL, macOS, or Linux without interrupting routine progress.

alanhuang168 Bundle

Generate, derive, optimize, and diagnose structured AI image prompts for Chinese guofeng donghua visuals, including dark xianxia / Renegade Immortal-like cold oppression, Battle Through the Heavens-like hot-blooded flame action, and general elegant guofeng animation. Use for character-forward ancient Chinese fantasy prompts, clear frontal faces, character consistency from reference images, two-person interactions such as hand-holding or eye contact, cinematic composition, lighting, palette, negative constraints, prompt rewrites, variants, or prompt audits across Codex, Claude, Grok, Kimi, MiniMax, Seedance, Workbuddy, Catpaw, and similar image-prompt workflows.

alessandrofare Bundle

A universal skill for Claude Code, Minimax Code, and any AI coding agent that accepts markdown skill files. Combines the Fable Brain behavioral prompt (8 modules that fix how a model responds), Karpathy's agent-orchestration mindset and workflow (program.md, AutoResearch, parallel tracks), and 50 production-ready mega-prompt templates invokable with `use template [NAME]`. Drop SKILL.md in as system instructions and the agent gets a working behavior, a working workflow, and 50 ready prompts.

alpayc Bundle

Diagnose and carry out the migration of an MCP server to the 2026-07-28 specification — the revision that made the transport stateless, formalized OAuth 2.1, and deprecated the logging, sampling, and roots capabilities. Use this skill whenever someone mentions migrating, upgrading, or checking an MCP server against the 2026-07-28 spec; asks whether their MCP server still works, is "ready", or will break; mentions the stateless transport model, Mcp-Session-Id, or OAuth 2.1 for MCP; or is debugging an MCP server that stopped working after a spec or SDK bump. Also use it when someone asks you to review or audit MCP server code for compatibility, even if they don't name the spec date.

alveeerahman Bundle

Design multi-agent and long-running agent systems: how many agents, supervisor vs swarm, what stops the loop, harness limits, evals, long-horizon prompts. Use when building or debugging one.

alveeerahman Bundle

Validate, test and score Claude Agent Skills: spec conformance, script testing, quality grade, security, token cost. Use when authoring or auditing a skill, or before uploading one to claude.ai.

ancienttwo Bundle

Execution-grade Xiaohongshu studio workflow for agencies and operators handling account launches, daily operations, multi-account delivery, and low-traffic diagnosis. Use when the agent needs to launch or run a 小红书 account, prepare client artifacts, continue an existing client workspace, or diagnose underperforming notes for 代运营 teams. Triggers: 小红书养号, 小红书起号, 代运营, 工作室运营, 多账号运营, 小红书日常运营, 低流量诊断, 账号诊断.

andreabozzo Bundle

Use when working with the Ares web scraper — an LLM-powered Rust tool that extracts structured data from websites using JSON Schemas. Covers library usage, CLI commands, REST API, schema creation, adding custom fetchers/cleaners/extractors, deployment, and contributing to the Ares codebase.

andrewavery7 Skill

Switch the current Claude Code session to Codex - transfers this conversation into a resumable Codex thread and opens the OpenAI Codex desktop app directly on it (VS Code panel or terminal on request), with a clickable model picker. Use when the user types /to-codex, says 'switch to codex', 'transfer to codex', 'move this to codex', 'continue in codex', or is running low on Claude tokens and wants to hand the work to Codex. Optional inline argument: a model shorthand (sol, terra, luna, 5.5, 5.4, mini) to skip the picker.

andrewbastin Skill

Explore and understand GitHub repositories using the gh CLI. Use when asked to browse repos, read files, search code, analyze issues/PRs/commits/releases, or understand codebases on GitHub. Replaces MCP-based GitHub tools with zero dependencies beyond gh.

angelamer Bundle

Analyze Python code, design and implement unit or integration tests, run them, and produce an evidence-based Markdown report. Use when any coding agent is asked to write, improve, execute, or report on Python tests, including projects that use unittest or pytest and workflows backed by a hosted API, an OpenAI-compatible endpoint, Ollama, or no external model.

anonymousyz Bundle

AI 上线否决卡：对“AI 功能或 Agent 准备上线”的方案做结构化红队评审，扫描 8 条否决条件， 按七个维度追问缺口，输出可带进评审会的否决卡。Use when the user wants to review, challenge, or red-team an AI launch, pilot, or deployment plan (AI 上线评审 / 就绪度检查 / 方案红队 / launch readiness review), or pastes a proposal claiming an AI system is ready for production.

anserinaecrotalariasagitallis995 Bundle

Design and build context-packet DAG pipelines — graph design, shell orchestration, MCP server integration, and programmatic TypeScript API. Use when creating AI agent workflows that pass context between nodes.

antreasantoniou Bundle

Establish, actively monitor, and operate project-agnostic collaboration control for multi-agent programmes, research campaigns, live experiments, long-running automation, consequential artifact production, incident recovery, evidence promotion, and cross-agent handoffs. Use when work needs ongoing agent support, a recurring independent watch, proactive anomaly or ambiguity reporting, explicit human authority, one writer per mutable surface, risk-scaled Agent Orchestra topology, transition journals, liveness proof, adversarial review, provenance, succession rules, or calibrated claim states across any project.

antreasantoniou Bundle

Operate a Git-backed Archivum as durable agent memory for projects, research, decisions, meetings, tasks, sources, and outputs. Use when the user asks to capture, find, connect, update, audit, or resume work in an Archivum, or to initialize an Archivum workspace. Do not use merely because a repository contains Markdown.

antreasantoniou Bundle

Install, operate, diagnose, and safely consume an owner-only macOS Bitwarden CLI lease that uses one native hidden-answer unlock interaction for a 24-hour process-continuity window. Use when repeated `bw unlock` prompts or agent handoffs disrupt work, when local agents need Bitwarden-backed credentials without persisting `BW_SESSION`, when checking whether the lease is usable, or when repairing the `io.github.antreasantoniou.bitwarden-lease` LaunchAgent. Bitwarden only; never substitute another credential store implicitly.

antreasantoniou Bundle

Preserve operational history and intent across agent sessions. Use when resuming work, recording a decision, preparing a destructive or bulk operation, verifying a deployment, correcting a claim, reconstructing a captured file version, or handing off unfinished work.

antreasantoniou Bundle

Reconcile recent Claude Code and Codex sessions for one project, create a local evidence packet, and maintain a small curated progress ledger both harnesses can read. Use for cross-agent handoffs, resuming work, surfacing conflicts, and recording verified decisions, evidence, blockers, next actions, and artifacts.

arch-dog Bundle

Convert a finalized script or script fragment into Seedance-oriented 10-15 second text storyboards and executable video prompts. Use when the user asks for 定稿剧本转分镜, 剧本片段转分镜, Seedance 分镜, 文字分镜, 视频提示词, 镜头提示词, AI 动画短剧分镜, or a Seedance-ready prompt table. Each storyboard may contain multiple internal shots. Do not use for plot expansion, multi-panel storyboard image generation, asset image creation, direct video generation, post-production, or generated-video review.

arvorco Bundle

Install, configure, operate, and troubleshoot the local-first Memento memory system. Use when a user asks to install Memento, connect Codex, Claude Code, OpenClaw, or another agent through MCP or CLI, initialize an Obsidian vault or document collection, ingest or synchronize sources, learn, search memory, inspect evidence, or diagnose retrieval quality.