本文へ移動
cccskills
無料GitHub で公開

mcp-manage

Use when checking capabilities or instructing the user on how to interact with specialized MCP servers like Unreal, Rhino, DaVinci, or TouchDesigner.

インストール方法を見る

含まれるファイル(1)

  • SKILL.md5.8 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

MCP Manage: Specialized Tool Orchestration

You are the authoritative skill for managing and utilizing the specialized Model Context Protocol (MCP) servers installed in this environment. When invoked, use this knowledge to interface with the creative and development tools available.

Available MCP Servers

  1. github (@modelcontextprotocol/server-github)

    • Capabilities: Read/write repositories, manage issues, handle PRs, search code.
  2. chrome-devtools (@modelcontextprotocol/server-puppeteer)

    • Capabilities: Automate Chrome/Chromium, run JS in browser, scrape dynamic pages, test web UI.
  3. bdb_unreal_mcp (Unreal Engine 5 C++/TS Bridge)

    • Capabilities: Execute Python or Blueprints in the UE editor, manipulate actors, manage levels, trigger renders.
    • Implementation: Cloned from the official ChiR24/Unreal_mcp including the native C++ McpAutomationBridge plugin.
  4. bdb_rhino_mcp & bdb_rhino_mcp_fallback (Rhino 3D MCP Servers)

    • Capabilities: Official McNeel RhinoMCP community server and GOLEM 3DMCP Python fallback with 105 tools. full NURBS, layers, Grasshopper, and rendering.
  5. bdb_davinci_mcp & bdb_davinci_mcp_fallback (DaVinci Resolve MCP Servers)

    • Capabilities: samuelgursky official Node server and hoyt-harness davinci-mcp-professional fallback. Automation of timelines, media pool, metadata, and color grading.
  6. bdb_grandma3_mcp (grandMA3 console)

    • Capabilities: lighting console control via WebAPI/OSC. Edit cue lists, patch fixtures, run macros.
  7. bdb_resolume_mcp (Resolume Arena REST)

    • Capabilities: Trigger clips, adjust deck parameters, clear layers, BPM sync.
  8. adobe_mcp & adobe_uxp_mcp (Adobe UXP/Scripting)

    • Capabilities: WebSocket bridge to native UXP plugins (Photoshop/Premiere) and python ExtendScript executor (Illustrator/Photoshop).
  9. bdb_blender_mcp & bdb_blender_mcp_fallback (Blender MCP Servers)

    • Capabilities: ahujasid socket-based BlenderMCP and djeada python server fallback. Scene inspection, material control, object manipulation, rendering.
  10. bdb_after_effects_mcp & bdb_after_effects_mcp_fallback (After Effects MCP Servers)

    • Capabilities: Dakkshin Node server and sunqirui1987 Go+ fallback. Compositions creation, layers styling, solid/shape preset renders, and ExtendScript triggers.
  11. bdb_vectorworks_mcp (Vectorworks CAD/BIM)

    • Capabilities: RAG-based search, command, and wiki indexer by togawamanabu. Guides the AI to write correct scripts and manage CAD data hierarchies.
  12. bdb_td_minddesigner & bdb_td_backup (TouchDesigner MCP Servers)

    • Capabilities: Pantani tdmcp MindDesigner TOP/CHOP node builder and 8beeeaaat fallback direct parameter controller.
  13. zavora_computer_use (Zavora Computer Use)

    • Capabilities: Take screenshots, move mouse, type text, press keys, read/write clipboard, window activation. Pre-compiled Rust native NAPI binaries.

How to use them

  • When the user asks to manipulate lighting, target bdb_grandma3_mcp or bdb_resolume_mcp.
  • When working on CAD or game environments, invoke bdb_rhino_mcp, bdb_unreal_mcp, or bdb_blender_mcp.
  • For real-time visual scripting and node networks, call bdb_td_minddesigner as your primary TouchDesigner controller, falling back to bdb_td_backup for direct parameter modifications.
  • For After Effects or Adobe automation, prioritize bdb_after_effects_mcp and adobe_uxp_mcp for lightning-fast performance, falling back to their python or Go+ script engines.
  • Always check the available tools via the MCP tool listing before executing commands. If a server is down, instruct the user to verify the mcp_config.json configuration and ensure the respective host applications are running with API/OSC/Lua/network plugins enabled.

Overview

mcp-manage is the authoritative router for all specialized BDB Model Context Protocol (MCP) servers, ensuring the AI agent selects the right tool for Unreal, Rhino, DaVinci, and others.

When to Use

  • Trigger: The user asks how to automate a specific creative app, or the agent needs to select the correct MCP server for a domain-specific task.
  • Exclude: Do not use for general file system operations or generic web searches.

Core Process

  1. Identify the target creative application (e.g., TouchDesigner, Rhino, DaVinci Resolve).
  2. Look up the primary MCP server and its fallback in the available list.
  3. Instruct the user to verify the application is running and the MCP config is correct if the server is unreachable.
  4. Call the appropriate server tool to execute the automation.

Common Rationalizations

RationalizationReality
"I can just write a Python script and run it externally."MCP tools provide direct WebSocket/native integration; external scripts lose context and safety bounds.
"I'll use the fallback server first since it's simpler."Always use the primary optimized MCP server (e.g., bdb_td_minddesigner) unless it fails.
"The server isn't responding, I'll just give up."You must verify if the host app (e.g., Unreal Engine) is actually open and running before failing.

Red Flags

  • Recommending generic OS scripts when a dedicated MCP server (like bdb_rhino_mcp) exists.
  • Failing to suggest checking mcp_config.json when a server connection fails.
  • Using a fallback server without first attempting the primary optimized server.

Verification

  • The correct MCP server was invoked based on the target application.
  • Fallbacks were only utilized if the primary server failed.
  • The agent verified the required host application was running.

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Use when manage multiple local CLI agents via tmux sessions (start/stop/monitor/assign) with cron-friendly scheduling.

日本語の概要は準備中です。原文の説明を表示しています。

hybridlabor-api/aos62026年10月12日 更新

Use when a hybrid memory system that provides persistent, searchable knowledge management for AI agents (Architecture, Patterns, Decisions).

日本語の概要は準備中です。原文の説明を表示しています。

hybridlabor-api/aos62026年10月12日 更新

Reference for how BDB structures autonomous software engineering work — the seven-node dispatcher graph (Architect, TechLead, UI/UX, Engineering, Media/EventTech, Reviewer, Shipping) that /startcycle-graph actually runs. Use when you need the high-level lifecycle framing without inventing your own process.

日本語の概要は準備中です。原文の説明を表示しています。

hybridlabor-api/aos62026年10月12日 更新

Tools are how AI agents interact with the world. A well-designed tool is the difference between an agent that works and one that hallucinates, fails silently, or costs 10x more tokens than necessary. This skill covers tool design from schema to error handling.

日本語の概要は準備中です。原文の説明を表示しています。

hybridlabor-api/aos62026年10月12日 更新

Harness patterns for coding agents — memory, permissions, context engineering, delegation, skills, hooks, bootstrap.

日本語の概要は準備中です。原文の説明を表示しています。

hybridlabor-api/aos62026年10月12日 更新

agenttrail: live map of a multi-agent build in the browser: which plan component is being worked on, by which agent or harness, what is done and what is stuck. Use when a multi-agent pipeline starts (/startcycle, /startcycle-graph, /teamwork-preview) or after a plan-canvas approve, when the user asks to see what the agents are doing ("live map", "build board", "who is running now"), or by yourself whenever a multi-agent run is under way — no user prompt needed.

日本語の概要は準備中です。原文の説明を表示しています。

hybridlabor-api/aos62026年10月12日 更新

hybridlabor-api のスキルをすべて見る

このスキルの問題を報告する