Turn any component site, asset library, or API into something agents can use — a registry entry, a skill, or a full MCP server. Load when the user says "add this site/library to Atelier" or wants to contribute a new integration.
Author and organize Neptune DXP server scripts and script projects via the MCP tools `list_script_projects`, `get_script_project`, `create_script_project`, `delete_script_project`, `get_server_script`, `save_server_script`, `delete_server_script`, `list_ungrouped_scripts`. Use when the user wants to write, read, create, edit, organize, or delete a server-side JS/TS script — including scripts that read or write table data (link the table via `entitySets`, use `entities.<name>`). Trigger phrases include "create a server script", "edit the script", "new script project", "group these scripts", "list the scripts", "delete the script", "insert data from a script", "add mock/test data", "read the table in the script". For running/testing a script see the run-server-script skill.
Execute a Neptune DXP server script directly via the MCP `run_server_script` tool. Use when the user wants to invoke, test, or trigger a specific server script by id without going through an HTTP API endpoint. Trigger phrases include "run that script", "execute the server script", "trigger the script", "test my script".
Run GPU workloads on Modal — training, fine-tuning, inference, batch processing. Zero-config serverless: no SSH, no Docker, auto scale-to-zero. Use when user says \"modal run\", \"modal training\", \"modal inference\", \"deploy to modal\", \"need a GPU\", \"run on modal\", \"serverless GPU\", or needs remote GPU compute.
Configure the Canvas Authoring MCP server for the current coauthoring session. USE WHEN "configure MCP", "set up MCP server", "MCP not working", "connect Canvas Apps MCP", "canvas-authoring not available", "MCP not configured", "set up canvas apps".
Using the Exorails SQL Server connector well: T-SQL specifics, TOP instead of LIMIT, schema-qualified names, and the read-only guarantee the connector enforces.
Guides creation of ChatGPT Apps with interactive widgets using OpenAI Apps SDK and MCP servers. Use when building ChatGPT custom apps with visual UI components, embedded widgets, or rich interactive experiences. Covers widget architecture, MCP server setup with FastMCP, response metadata, and Developer Mode configuration. NOT when building standard MCP servers without widgets (use building-mcp-servers skill instead).
Build Model Context Protocol (MCP) servers in C#/.NET against the current ModelContextProtocol 2.x NuGet packages. Helps with cases the model gets wrong without guidance — stale versions (0.x preview or 1.x-era defaults), the v2 stateless-by-default HTTP flip, the 2026-07-28 spec deprecations (roots/sampling/logging), MCP Apps and Tasks extension packages, elicitation URL mode, per-session HTTP wiring, OAuth and reverse-proxy deploy specifics, and debugging MapMcp / STDIO / Streamable-HTTP errors. Also covers STDIO and Streamable HTTP transports (SSE is deprecated), tools, prompts, resources, completions, and a basic .NET MCP client. Trigger when the user says or implies any .NET MCP server work: ModelContextProtocol, McpServerTool, MapMcp, WithStdioServerTransport, "MCP server in C#", "MCP tool in dotnet", "expose this as MCP", or names a primitive (prompt/resource/elicitation/MCP App) in a .NET context. Skip for MCP work in other languages.
Manage a server through the ASC daemon CLI - install and control applications (Docker and native), view logs, monitor resources, create backups. Use when the user asks to install/start/stop apps on a server, check server or app status, read logs, back up an app, or troubleshoot a server managed by AdminService.Cloud (asc).
Start a local Microcks server with Docker Compose for API mocking and testing. Use when a developer needs to run Microcks locally to mock REST APIs from OpenAPI specifications. Microcks is a CNCF open-source tool for API mocking and testing.
Controls a running iOS, iPad, or Apple Watch Simulator via the serve-sim CLI (npx serve-sim) and streams it into the host agent's preview pane. Use whenever the user wants an AI agent to view or drive an Apple Simulator — streaming to preview, taps at normalized coordinates, multi-touch gestures, hardware buttons, rotation, memory warnings, CoreAnimation debug, synthetic camera injection, or media drag-drop. Triggers include "serve-sim", "iOS simulator", "Apple simulator", "iPad simulator", "Apple Watch simulator", "stream the simulator", "show the simulator in preview", "view the simulator here", "open simulator in preview", "simulator gestures", "tap on the simulator", "rotate the simulator", "inject camera feed", or any request to drive or display an Apple Simulator visually. Do NOT use for Android emulators, building/installing an iOS app (use xcodebuild), booting a simulator from scratch (use xcrun simctl boot), in-app React Native runtime debugging (use rn-debugger), or real iOS hardware.
Настройка серверного прокси: создание mixed inbound (SOCKS5+HTTP) на 127.0.0.1:1080 в существующей панели 3X-UI (со sniffing — иначе domain-правила для серверных программ мертвы), маршрутизация через VPN-outbound по модели «золотая середина» (private→direct, реклама→block, geoip:ru + category-ru + regex→direct, остальное→upstream), systemd drop-in override для x-ui (КРИТИЧНО — без него self-loop через /etc/environment ломает панель), запись socks5h://-переменных в /etc/environment. Финальный smoke-test (api.anthropic.com через прокси, ya.ru напрямую). Цель — дать программам на сервере (боты, npm, pip, curl, git) доступ к заблокированным API через VPN, без касания UI/настроек каждой программы отдельно. Триггеры: «настрой прокси на сервере», «бот не видит Anthropic API», «npm не качает», «pip не работает», «бот возвращает 403 от Anthropic», «git push с сервера не работает», «curl таймаутит на api.anthropic.com», «дай серверу доступ к заблокированным API». НЕ для установки панели — `/setup-vpn-panel`. НЕ для добавления клиентов устройствам — `/configure-vpn-routing`. НЕ для генерации клиентских конфигов — `/generate-client-config`.
Set up, configure, build, analyze, and operate Minecraft Java Edition servers for 1.21.x across Paper, Purpur, Folia, Velocity networks, and modded (Fabric/NeoForge) deployments. Use this for Paper/Purpur/Folia/Velocity server building, plugin discovery/download/install planning from Hangar, Modrinth, GitHub, SpigotMC, or BukkitDev, plugin marketplace/source review, server zip/folder analysis, server archetype design (lobby, SMP, creative, minigames, proxy network), plugin compatibility/configuration planning, performance tuning, backups, incidents, Docker/Pterodactyl patterns, and security hardening. Use for Java server infrastructure and plugin operations, not for writing plugin or mod code.
Builds, deploys, manages, debugs, configures, and optimizes serverless applications on AWS using Lambda, API Gateway, Step Functions, EventBridge, and SAM/CDK. Covers cold starts, CORS debugging, event source mappings, troubleshooting, concurrency, SnapStart, Powertools, function URLs, EventBridge Scheduler, Lambda layers, and production readiness. Triggers on mentions of Lambda, API Gateway, Step Functions, SAM templates, CDK serverless stacks, DynamoDB stream triggers, SQS event sources, cold starts, timeouts, 502/504 errors, throttling, concurrency, CORS, Powertools, or any event-driven architecture on AWS, even without the word "serverless." Does not apply to EC2, ECS/Fargate containers, or Amplify hosting.
Builds remote MCP (Model Context Protocol) servers on Cloudflare Workers with tools, OAuth authentication, and production deployment. Generates server code, configures auth providers, and deploys to Workers.
Serve RpcGroup contracts with Effect's RpcServer — handler layers (RpcGroup.toLayer/toLayerHandler), protocol layers (HTTP, WebSocket, TCP, stdio, worker), RpcSerialization, server middleware, streaming results, interruption and shutdown semantics, RpcTest. Use when implementing the server side of an Effect RPC API, mounting RPC on an HttpRouter or existing HTTP app, choosing a wire format, implementing RpcMiddleware, handling client aborts, or testing RPC handlers.
Validate MCP (Model Context Protocol) server tools end-to-end after code changes. Use this skill whenever the user is building, modifying, or deploying an MCP server and needs to verify that all registered tools actually work — not just that the server starts. Triggers on mentions of "MCP server", "MCP tools", "tool validation", "tools/list", "tools/call", or any task where Docker rebuild + health check is followed by a need to confirm tool behavior. Also triggers when the user wants an automated feedback loop that detects tool failures, fixes code, redeploys, and re-tests until green before raising a PR. Do NOT use for general API testing, unit tests, or non-MCP servers.
Plain text files in a repository that tell a coding agent how the project works: commands to run, conventions to follow and things to avoid. CLAUDE.md, AGENTS.md, cursor rules and skills are the common kinds.
CLAUDE.md or AGENTS.md?
CLAUDE.md is read by Claude Code. AGENTS.md is an open format that Codex, Cursor and other agents read. Many projects keep one and point the other at it.
What is a skill?
A folder with a SKILL.md that describes one capability, such as filling PDFs or reviewing code. The agent loads it only when the task calls for it.
Can I search my own team's files too?
Your agents already can, over MCP, limited to the files you're allowed to read. Searching them from this page is coming.