@hybridlabor-api/bdb-antigravity-skills 1.0.2 → 1.0.4

package/README.md

@@ -1,8 +1,8 @@
 # 🚀 BDB DEV - Optimized Antigravity Skills
 
-Welcome to the **BDB DEV Antigravity Skills & MCP Configuration** repository. This project serves as the backbone of our creative development ecosystem, supercharging AI agents with highly specialized capabilities tailored specifically for the event and creative technology industry.
+Welcome to the **BDB DEV Antigravity Skills & MCP Configuration** repository. This project serves as the backbone of our creative and full-stack development ecosystem, supercharging AI agents with highly specialized capabilities tailored for the event and creative technology industry, as well as general-purpose software engineering.
 
-By unifying 140 ultra-optimized skills and integrating state-of-the-art Model Context Protocol (MCP) servers, this repository transforms the Antigravity workflow into an unstoppable force for show control, 3D development, and interactive experiences.
+By unifying 140 ultra-optimized skills and integrating state-of-the-art Model Context Protocol (MCP) servers, this repository transforms the Antigravity workflow into an unstoppable force for show control, 3D development, interactive experiences, and robust full-stack applications.
 
 ---
 
@@ -10,7 +10,13 @@ By unifying 140 ultra-optimized skills and integrating state-of-the-art Model Co
 
 We started with a massive pool of over 1,400 raw AI skills. After rigorous testing, filtering, and refinement, we’ve distilled them down to a hyper-curated set of **140 Optimized Antigravity Skills**. 
 
-These skills are precision-engineered for the **BDB DEV daily workflow**, ensuring our agents waste no time on redundant tasks and instead operate with maximum agency and context awareness. Whether it's spinning up a Next.js App Router boilerplate, structuring complex databases, or auditing UI/UX patterns, these 140 skills cover every aspect of modern full-stack and creative technology development.
+These skills are precision-engineered for how we believe AI should be integrated into a professional daily workflow, ensuring agents waste no time on redundant tasks and instead operate with maximum agency and context awareness.
+
+### 💻 Beyond Events: Normal Software & App Coding
+While heavily optimized for the creative tech industry, these skills are deeply rooted in core software engineering. They include specialized workflows for:
+- **Full-Stack Development**: Spinning up Next.js App Router boilerplates, building scalable Node.js microservices, and crafting interactive frontends.
+- **App Development**: Architecting databases, designing REST APIs, and building standard web and mobile applications from scratch.
+- **Design & Quality Assurance**: Auditing UI/UX patterns, enforcing clean code principles, and setting up strict CI/CD pipelines.
 
 ---
 
@@ -22,43 +28,90 @@ BDBrainstorm combines multi-agent brainstorming, the `/grill-me` slash command,
 
 ---
 
-## 🔌 Custom Whitelabeled MCP Integrations
+## 🔌 Custom MCP Integrations
+
+The true game-changer of this repository lies in our **Custom MCP (Model Context Protocol) Integrations**. These bridges allow our Antigravity agents to natively read, write, and execute commands within the industry's leading creative software.
+
+### 🎛️ TouchDesigner
+Integrated directly into TouchDesigner via the **Pantani/tdmcp** (MindDesigner) bridge and utilizing **8beeeaaat/touchdesigner-mcp** as a fallback, agents can construct real TouchDesigner node networks via natural language. They can manipulate operators, patch CHOPs/TOPs, and automate complex node routing inside the visual programming environment.
+
+### 🎮 Unreal Engine
+Built on a hybrid foundation of **Unreal Engine 5.8 native APIs** and the **gimmeDG** toolset, this MCP allows agents to interact directly with UE5 projects. From scene generation and asset configuration to complex Blueprint logic mapping, this integration turns agents into bona fide Technical Artists.
+
+### 📐 Rhino 3D & Grasshopper
+Using **mcneel/RhinoMCP** and **TheKingHippopotamus/GOLEM-3DMCP-Rhino-**, agents can now create and manipulate 3D geometry in Rhino 8. This extends to controlling Grasshopper definitions, tweaking parameters, and generating complex parametric 3D models directly from prompt instructions.
+
+### 🎬 DaVinci Resolve
+Powered by **samuelgursky/davinci-resolve-mcp** and **hoyt-harness/davinci-mcp-professional**, this integration gives agents the ability to manipulate timelines, organize media pools, and execute complex Fusion composites via external scripting in Resolve Studio.
 
-The true game-changer of this repository lies in our **Custom Whitelabeled MCP (Model Context Protocol) Integrations**. These bridges allow our Antigravity agents to natively read, write, and execute commands within the industry's leading creative software.
+### 🧊 Blender
+Using community servers like **ahujasid/blender-mcp**, agents can script Blender Python (`bpy`) operations directly. This covers everything from mesh generation and material manipulation to camera automation and rendering pipelines.
 
-### 🎮 BDB Unreal
-Built on a hybrid foundation of **Unreal Engine 5.8 native APIs** and the **gimmeDG** toolset, the BDB Unreal MCP allows agents to interact directly with UE5 projects. From scene generation and asset configuration to complex Blueprint logic mapping, this integration turns agents into bona fide Technical Artists.
+### ✨ After Effects
+Integrated via servers like **Dakkshin/after-effects-mcp** and **sunqirui1987/ae-mcp**, agents can generate compositions, manipulate layers, set keyframes, and write custom ExtendScript expressions, automating motion graphics pipelines.
 
-### 🎬 BDB Resolume
-Driven by the **tortillaguy** project and the **Resolume 7.26 native integration**, our agents can now control live visual environments. They can programmatically structure compositions, map OSC/MIDI routes, and sequence layers, bringing AI-driven generative logic to live VJ setups and permanent installations.
+### 🏗️ Vectorworks
+Through early implementations like **vectorworks-mcp** connecting via the C++ SDK plugin, agents are paving the way for automated drafting, BIM parameter adjustments, and CAD automation within Vectorworks 2025.
 
-### 💡 BDB grandMA3
-Lighting programming takes a massive leap forward with the BDB grandMA3 MCP. Powered by a specialized **Python/Lua bridge** built on top of the **Pahegi/ma3-mcp** foundation, agents can now read showfiles, generate macros, parse patch structures, and script complex lighting sequences without leaving their terminal environment.
+### 💡 grandMA3 & Resolume
+- **grandMA3**: Powered by a specialized **Python/Lua bridge** (Pahegi/ma3-mcp), agents can read showfiles, generate macros, and patch fixtures.
+- **Resolume**: Driven by the **tortillaguy** project and the **Resolume 7.26 native integration**, agents can structure compositions, map OSC/MIDI routes, and sequence layers.
 
 ---
 
 ## ⚡ Workflow Transformation
 
-This setup completely redefines the **Antigravity workflow for event and creative development**.
+This setup completely redefines the **Antigravity workflow**. 
 
-By combining the **140 Optimized Skills** with our **Whitelabeled MCPs**, we collapse the gap between software development and show production. An agent can now write a Python microservice, generate the web-based UI for it, write a Lua script for grandMA3 to receive the data, and configure the Unreal Engine scene to visualize the output—**all in a single, autonomous flow.**
+By combining the **140 Optimized Skills** with our **Custom MCPs**, we collapse the gap between software development and production software. An agent can now write a Python microservice, generate the web-based UI for it, write a Lua script for grandMA3 to receive the data, generate a TouchDesigner visual component, and configure an Unreal Engine scene to visualize the output—**all in a single, autonomous flow.**
 
 ---
 
 ## 🛠️ Installation
 
-You can have an Antigravity agent install these skills by pointing it to the `agent.md` file, or you can do it manually by running the provided installer script:
+We support multiple installation methods depending on your environment. The installer will safely back up any existing skills, deploy the curated global/workspace skills, and configure your MCP settings.
+
+### Option 1: Using NPM / NPX (Recommended for Node users)
+
+You can run the installer directly without downloading the repository using `npx`:
+
+```bash
+npx @hybridlabor-api/bdb-antigravity-skills
+```
+
+Or install it globally via npm:
+
+```bash
+npm install -g @hybridlabor-api/bdb-antigravity-skills
+bdb-skills
+```
+
+### Option 2: Using Homebrew (macOS)
+
+If you are on a Mac and prefer Homebrew, you can install the package directly:
+
+```bash
+brew tap hybridlabor-api/bdb-skills
+brew install bdb-skills
+bdb-skills
+```
+
+### Option 3: Manual Shell Script
+
+Clone the repository and run the installer script:
 
 ```bash
+git clone https://github.com/hybridlabor-api/bdb-dev-optimized-antigravity-skills.git
+cd bdb-dev-optimized-antigravity-skills
 chmod +x installer.sh
 ./installer.sh
 ```
 
-**The script will automatically:**
+**The installer script will automatically:**
 1. Back up your existing global and workspace skills safely.
 2. Deploy the new curated global config skills.
 3. Install the workspace-specific agent skills.
 4. Copy the customized `GEMINI.md` to `~/.gemini/GEMINI.md`.
 
 ---
-*Elevate your agency. Dominate the show.*
+*Elevate your agency. Dominate the workflow.*

package/assets/touchdesigner/bootstrap.py

@@ -0,0 +1,106 @@
+"""One-paste TouchDesigner bridge bootstrap (no clone, no Preferences).
+
+Paste this single line into the Textport (Dialogs -> Textport and DATs) and the
+bridge installs itself and starts:
+
+    import urllib.request; exec(urllib.request.urlopen("https://github.com/Pantani/tdmcp/raw/v0.11.0/td/bootstrap.py").read().decode())
+
+That release-pinned snippet downloads just the bridge modules to
+~/tdmcp-bridge/modules, puts them on sys.path for this session, and runs
+install.run() -> a tdmcp_bridge on port 9980.
+
+Requires the GitHub repo (or release zip) to be reachable. If your repo is
+private, point REPO_ZIP at a public release asset, or use `install-bridge`
+(`node dist/index.js install-bridge`) from a local checkout instead.
+"""
+
+import io
+import os
+import stat
+import sys
+import zipfile
+import urllib.request
+
+REPO_ZIP = "https://github.com/Pantani/tdmcp/archive/refs/tags/v0.11.0.zip"
+DEST = os.path.expanduser("~/tdmcp-bridge")
+_MARKER = "/td/modules/"
+_SKIP_RUN_ENV = "TDMCP_BOOTSTRAP_SKIP_RUN"
+
+
+def _is_symlink(info):
+    return stat.S_ISLNK((info.external_attr >> 16) & 0o170000)
+
+
+def _safe_module_path(name, modules_dir):
+    idx = name.find(_MARKER)
+    if idx == -1:
+        return None
+
+    rel = name[idx + len(_MARKER):].replace("\\", "/")
+    if not rel or rel.endswith("/"):
+        return None
+
+    parts = rel.split("/")
+    if (
+        rel.startswith("/")
+        or rel.startswith("\\")
+        or (len(parts[0]) >= 2 and parts[0][1] == ":")
+        or any(part in ("", ".", "..") for part in parts)
+    ):
+        raise RuntimeError("[tdmcp] Refusing unsafe archive entry: %s" % name)
+
+    root = os.path.realpath(modules_dir)
+    target = os.path.realpath(os.path.join(modules_dir, *parts))
+    if target != root and not target.startswith(root + os.sep):
+        raise RuntimeError("[tdmcp] Refusing archive entry outside modules: %s" % name)
+
+    return target
+
+
+def fetch_modules(repo_zip=REPO_ZIP, dest=DEST):
+    """Download the repo zip and extract only its td/modules tree into dest/modules."""
+    modules_dir = os.path.join(dest, "modules")
+    try:
+        data = urllib.request.urlopen(repo_zip, timeout=30).read()
+    except Exception as exc:  # noqa: BLE001 - surface a friendly hint in the Textport
+        raise RuntimeError(
+            "[tdmcp] Could not download the bridge from %r (%s). If the repo is "
+            "private, use a public release asset or the local 'install-bridge' "
+            "command instead." % (repo_zip, exc)
+        )
+
+    zf = zipfile.ZipFile(io.BytesIO(data))
+    os.makedirs(modules_dir, exist_ok=True)
+    extracted = 0
+    for info in zf.infolist():
+        name = info.filename
+        if name.endswith("/"):
+            continue
+        target = _safe_module_path(name, modules_dir)
+        if target is None:
+            continue
+        if _is_symlink(info):
+            raise RuntimeError("[tdmcp] Refusing symlink archive entry: %s" % name)
+        os.makedirs(os.path.dirname(target), exist_ok=True)
+        with zf.open(name) as src, open(target, "wb") as out:
+            out.write(src.read())
+        extracted += 1
+
+    if extracted == 0:
+        raise RuntimeError("[tdmcp] Downloaded archive had no td/modules — wrong REPO_ZIP?")
+    print("[tdmcp] bridge modules -> %s (%d files)" % (modules_dir, extracted))
+    return modules_dir
+
+
+def run(repo_zip=REPO_ZIP, dest=DEST, port=9980):
+    modules_dir = fetch_modules(repo_zip, dest)
+    if modules_dir not in sys.path:
+        sys.path.insert(0, modules_dir)
+    from mcp import install
+
+    return install.run(port=port, modules_dir=modules_dir)
+
+
+# Running via exec(urlopen(...).read()) or as a script kicks off the install.
+if os.environ.get(_SKIP_RUN_ENV) != "1":
+    run()

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@hybridlabor-api/bdb-antigravity-skills",
-  "version": "1.0.2",
+  "version": "1.0.4",
   "description": "Optimized Antigravity skills and MCP pack for BDB DEV",
   "main": "installer.sh",
   "bin": {
@@ -14,6 +14,7 @@
     "mcp_config.json",
     "GEMINI.md",
     "skilloverview.html",
+    "assets/",
     "skills/"
   ],
   "author": "BDB DEV",