Use whenever the user wants to find, rank, or shortlist directories and registries where they can submit or list an MCP server (Model Context Protocol server) — to get backlinks, referral traffic, and discovery by agent builders. Triggers on "where do I list my MCP server", "best MCP directories", "MCP registries to submit to", "get my MCP server discovered", or "pull submission details for these MCP-directory domains", even when described indirectly (we built an MCP server, where do we publish it). Drives the ServiceGraph API (api.servicegraph.co) — a catalog of 1,000+ product directories enriched with Domain Rating, backlinks, and organic traffic. Defer to find-ai-directories for general AI-tool / AI-agent / agent-skill listings, and to find-product-directories for general SaaS/software launches. Skip building an MCP server or asking how MCP works (DIY), finding a firm to build one (use find-ai-consultancy / find-software-developer), and MCP link-building *services*.
Set up, operate, tune, and troubleshoot Minecraft Java 26.x and legacy 1.21.x servers across Paper, Purpur, Folia, Velocity, Fabric, and NeoForge. Use for infrastructure, backups, proxies, and live operations, not plugin or mod development.
Health-check the connected MCP servers and scout new ones: the safe pre-flight before integration work, plus a connector finder. Read-only by default. Triggers: "/mcp", "/mcp health", "/mcp discover <need>", "which MCP servers are alive", "find an MCP for <X>".
Self-hosted deployment platform. Use when (a) deploying or operating apps and infrastructure on a remote server; (b) working in a repo that contains `.servel/` or `servel.yaml` and the user references the app, the deployment, production, the live container, or the env/logs/secrets/routing of the running version; (c) debugging a deployed app — "why is prod 500", "show me the logs", "tail the logs", "what env did the container actually get", "the live app is crashing", "production is slow", "exec into the container", "my deployment is down", "the deployed version misbehaves"; (d) any infrastructure task — postgres, redis, supabase, mongodb, libsql, mysql, mongodb-ha, postgres-ha, backup, restore, registry (ghcr / gitlab / self-hosted), domains, SSL, traefik routing, secrets, alerts, capacity, analytics, visitors, rebalance, node management, access control, bastion, tunnel, port-forward, swap / zram / OOM, dev mode (link-env / link-infra / pull prod env to dev), CI/CD, auto-update / rolling upgrade, cross-node migration (servel move), volumes, audit, IP bans, security checkpoint, visitor IP / X-Real-Ip / CF-Connecting-IP. Enforces dogfooding: ALWAYS prefer servel commands over raw docker / ssh / rsync / docker-compose for any operation servel can perform.
Run OpenMed's Model Context Protocol (MCP) server so coding agents (Claude Code, Codex) and chat clients can call clinical NER, PII extraction, and de-identification as tools, on-device. Use when the user wants to add OpenMed to an agent's MCP config, expose de-id/NER as MCP tools, run an MCP server over stdio or Streamable HTTP, give Claude/Codex access to OpenMed, or containerize the MCP server. Covers the mcp extra, create_mcp_server, the 7 tools (openmed_analyze_text, openmed_extract_pii, openmed_deidentify, openmed_list_models, openmed_list_pii_languages, openmed_loaded_models, openmed_unload_model), the resources and prompts, stdio vs streamable-http transports, ServiceRuntime env config, and MCP client config snippets.
Перенос Docker-инфраструктуры со старого VPS на новый. 4 стратегии по downtime/cost: live (logical replication, 30-120 сек), backup-restore (10-60 мин), rsync-incremental (5-15 мин), blue-green (0 downtime, 2x cost). Pre-checklist + cutover (DNS, зеркало proxy_pass на старом — ноль split-brain) + verify (row counts, healthchecks, рабочее место оператора). Rollback явный. Триггеры: «переезд на новый сервер», «migrate server», «сменить VPS», «переместить инфру», «новый провайдер», «сменить хостинг». НЕ для bootstrap (это bootstrap-new-server); НЕ для приведения хаоса в порядок (cleanup-existing-server).
Configures the serena MCP server for the current project so it works identically in the Claude Code CLI and in "Code" inside the Claude Desktop app. Writes a committed self-locating wrapper script plus a machine-local .mcp.json. Invoke only when the user explicitly runs the `/setup-serena-mcp` slash command — do not trigger on general phrases like "set up serena" or "add an MCP server" unless that exact slash command is used.
Start the OCI Database Management MCP server locally so Claude can call DBM tools. This skill wires the MCP server into Claude's tool context, making all 24 DBM tools available. Use this skill when a user says "start the database management server", "connect to my database tools", "enable database management tools", or after oci-setup completes successfully. This skill must run before any oci-dbm-* operational skills (diagnose, performance, jobs, sql-tuning). It is the gateway — nothing else works without it.
Add a new MCP server to all AI tool configs. Use when user says "add MCP server", "new MCP server", "create MCP", "add tool to MCP", or wants to register a new MCP server across Claude Code, Codex, Cursor, and OpenCode.
Add a new MCP server to all AI tool configs. Use when user says "add MCP server", "new MCP server", "create MCP", "add tool to MCP", or wants to register a new MCP server across Claude Code, Codex, Cursor, and OpenCode.
Migrates an MCP server with interactive widgets from the OpenAI Apps SDK (window.openai, text/html+skybridge) to the MCP Apps standard (@modelcontextprotocol/ext-apps), covering server-side and client-side changes.
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.