dcc-mcp-core

dcc-mcp-core — DCC MCP Ecosystem Foundation

The foundational library enabling AI assistants to interact with Digital Content Creation (DCC) software through the Model Context Protocol (MCP).

Quick Decision Guide — Use the Right API

What This Library Does

Installation

For agent-side DCC control, install the published dcc-mcp Skill and use its CLI workflow; the Python package below is for adapters and embedded runtimes:

openclaw skills install @loonghao/dcc-mcp
# Direct ClawHub CLI:
npx --yes [email protected] install @loonghao/dcc-mcp

Use dcc-mcp-creator only for a complete adapter/runtime, and dcc-mcp-skills-creator only for a DCC-specific Skill package.

pip install dcc-mcp-core
# Python 3.7-3.14, zero runtime dependencies

Local Dependency Maintenance

Use the repository-pinned vx toolchain for Rust dependency refreshes and CI parity:

vx --version  # CI pins loonghao/[email protected]
vx cargo update
vx cargo tree -d
vx cargo build --workspace --all-targets --timings

Review duplicate dependency output before editing manifests, and keep generated lockfile changes only when they are part of the intended dependency refresh.

Core Patterns

Pattern 1: Skills-First — one-call MCP server (recommended)

import os
from dcc_mcp_core import create_skill_server, McpHttpConfig

os.environ["DCC_MCP_MAYA_SKILL_PATHS"] = "/opt/my-skills"

# One call: creates registry + dispatcher + catalog + discovers skills + server
server = create_skill_server("maya", McpHttpConfig(port=8765))
handle = server.start()
print(f"Maya MCP server: {handle.mcp_url()}")

# Agents connect and use on-demand skill discovery:
# → search_tools(query="bevel") or search_skills(query="modeling")
# → get_skill_info(skill_name="maya-bevel") to inspect schemas
# → load_skill("maya-bevel") only when selected
# → tools/call maya_bevel__bevel to execute
# Do not treat the first tools/list page as complete; follow nextCursor if listing.
handle.shutdown()

Pattern 2: Return structured results (always use factories)

from dcc_mcp_core import success_result, error_result, from_exception

# All actions should return ActionResultModel
def my_action(params):
    try:
        result = do_work(params)
        return success_result(
            f"Created {result['name']}",
            prompt="Object created. You can now modify its properties.",
            object_name=result["name"],
        )
    except Exception as e:
        return from_exception(str(e), message="Action failed")

Pattern 3: Validate action inputs

import json
from dcc_mcp_core import ToolValidator, error_result

schema = json.dumps({
    "type": "object",
    "required": ["name", "radius"],
    "properties": {
        "name": {"type": "string", "maxLength": 64},
        "radius": {"type": "number", "minimum": 0.001},
    },
})
validator = ToolValidator.from_schema_json(schema)
ok, errors = validator.validate(json.dumps(params))
if not ok:
    return error_result("Invalid parameters", "; ".join(errors))

Pattern 4: Connect to a running DCC via IPC

from dcc_mcp_core import DccLinkFrame, IpcChannelAdapter, success_result, error_result

# Connect to a DCC process via named pipe / Unix domain socket
channel = IpcChannelAdapter.connect("dcc-mcp-maya-12345")
try:
    # Send a Call frame and receive the reply
    channel.send_frame(DccLinkFrame(msg_type=1, seq=1, body=b'{"method":"execute_python","params":"cmds.sphere()"}'))
    reply = channel.recv_frame()  # DccLinkFrame
    if reply.msg_type == 2:  # Reply
        return success_result(reply.body.decode())
    else:
        return error_result("DCC call failed", reply.body.decode())
finally:
    channel.shutdown() if hasattr(channel, 'shutdown') else None

Pattern 5: Build a DCC adapter with DccServerBase

from pathlib import Path
from dcc_mcp_core import DccServerBase, DccServerOptions

class BlenderMcpServer(DccServerBase):
    def __init__(self, port: int = 8765, **kwargs):
        opts = DccServerOptions.from_env(
            "blender",
            Path(__file__).parent / "skills",
            port=port,
            **kwargs,
        )
        super().__init__(options=opts)

    def _version_string(self) -> str:
        import bpy
        return bpy.app.version_string

# All skill methods, hot-reload, gateway are ready:
server = BlenderMcpServer(port=8765)
server.register_builtin_actions()
handle = server.start()
print(f"MCP: {handle.mcp_url()}")

Pattern 6: Watch skills for live reload

from dcc_mcp_core import SkillWatcher

watcher = SkillWatcher(debounce_ms=300)
watcher.watch("/my/dev/skills")  # immediate load + start watching

# Get always-up-to-date snapshot
current_skills = watcher.skills()  # -> List[SkillMetadata]

Pattern 7: ActionDispatcher with handlers

import json
from dcc_mcp_core import ToolRegistry, ToolDispatcher

reg = ToolRegistry()
reg.register("create_sphere",
    input_schema=json.dumps({"type": "object", "required": ["radius"],
                              "properties": {"radius": {"type": "number", "minimum": 0.0}}}))

dispatcher = ToolDispatcher(reg)
dispatcher.register_handler("create_sphere", lambda params: {"created": True, "r": params["radius"]})

# Introspect handlers
dispatcher.has_handler("create_sphere")  # True
dispatcher.handler_count()               # 1
dispatcher.handler_names()               # ["create_sphere"]
dispatcher.remove_handler("create_sphere")  # True

result = dispatcher.dispatch("create_sphere", json.dumps({"radius": 2.0}))
# result == {"action": "create_sphere", "output": {"created": True, "r": 2.0}, "validation_skipped": False}

Pattern 8: DCC main-thread safety

Most DCC applications (Maya, Blender, Houdini) require scene API calls on their main thread. For Python adapters, prefer the public host bridge/dispatcher stack and pass it through DccServerOptions before skills are loaded. Low-level DeferredExecutor details are covered in docs/guide/dcc-thread-safety.md.

from pathlib import Path
from dcc_mcp_core import DccServerBase, DccServerOptions, HostExecutionBridge, InProcessCallableDispatcher

dispatcher = InProcessCallableDispatcher()  # replace with the DCC UI-thread dispatcher
bridge = HostExecutionBridge(dispatcher=dispatcher)
opts = DccServerOptions.from_env("maya", Path("skills"), execution_bridge=bridge)
server = DccServerBase(options=opts)
handle = server.start()

Pattern 9: Write skill scripts with skill_entry

from dcc_mcp_core.skill import skill_entry, skill_success, skill_error, skill_exception

@skill_entry
def create_sphere(radius: float = 1.0, name: str = "sphere") -> dict:
    import maya.cmds as cmds
    obj = cmds.polySphere(r=radius, n=name)[0]
    return skill_success(
        f"Created sphere '{obj}' with radius {radius}",
        prompt="You can now adjust properties or add materials.",
        object_name=obj,
        radius=radius,
    )

Pattern 10: Launch an isolated DCC child

Use child-only environment overrides instead of mutating os.environ when multiple artist and automation sessions share a machine:

from dcc_mcp_core import PyDccLauncher

launcher = PyDccLauncher()
info = launcher.launch(
    name="nuke-mcp",
    executable="Nuke15.2",
    args=["--disable-nuke-frameserver", "project.nk"],
    environment={
        "NUKE_DISABLE_FRAMESERVER": "1",
        "DCC_MCP_NUKE_PORT": "0",
    },
    working_directory="/projects/solar-system",
)

Creating a Custom Skill (Zero Python Code)

# 1. Create directory structure
mkdir -p my-tool/scripts/

# 2. Write SKILL.md (name is required, follows agentskills.io spec)
cat > my-tool/SKILL.md << 'EOF'
---
name: my-tool
description: "My custom DCC automation tools. Use when automating scene setup or batch operations."
compatibility: "python>=3.7"
allowed-tools: "python"
metadata:
  dcc-mcp:
    dcc: maya
    version: "1.0.0"
    layer: example
    tags: ["automation", "custom"]
    tools: tools.yaml
---

# My Tool

Automation scripts for Maya workflow optimization.
EOF

# 3. Add the sibling tool declaration referenced by metadata.dcc-mcp.tools
cat > my-tool/tools.yaml << 'YEOF'
tools:
  - name: list_selected
    description: List selected objects in the Maya scene.
    input_schema:
      type: object
      properties: {}
    read_only: true
    idempotent: true
    source_file: scripts/list_selected.py
YEOF

# 4. Add a script
cat > my-tool/scripts/list_selected.py << 'PYEOF'
#!/usr/bin/env python3
"""List selected objects in the Maya scene."""
import json

result = {"selected": ["pSphere1", "pCube1"], "count": 2}
print(json.dumps(result))
PYEOF

# 5. Use it
export DCC_MCP_SKILL_PATHS="$(pwd)/my-tool"
python -c "
from dcc_mcp_core import scan_and_load
skills, _ = scan_and_load(dcc_name='maya')
print(f'Loaded: {[s.name for s in skills]}')
# Action: my_tool__list_selected
"

Architecture Overview

┌─────────────────────────────────────────────────────┐
│                   Python Layer                       │
│  dcc_mcp_core/__init__.py  →  _core (PyO3 cdyll)   │
│  380+ public symbols re-exported from Rust core      │
│  + Pure-Python: DccServerBase, DccServerOptions,     │
│    gateway election, hot-reload, factory, helpers    │
└──────────────────────┬──────────────────────────────┘
                       │ PyO3 bindings
┌──────────────────────▼──────────────────────────────┐
│        Rust Workspace (47 members total)             │
│        46 functional crates + workspace-hack         │
│  naming → models → actions → skills → protocols     │
│  gateway/http-types/http-server/http-py/http         │
│  host → transport → process → sandbox → telemetry   │
└─────────────────────────────────────────────────────┘

Environment Variables

Key Files in This Repository

Supported DCC Software

• Autodesk Maya — MEL/Python scripting (dcc: maya)
• Blender — Python API (dcc: blender)
• SideFX Houdini — HScript/Python (dcc: houdini)
• Autodesk 3ds Max — MaxScript/Python (dcc: 3dsmax)
• Any DCC — Generic Python wrapper (dcc: python)

Related Projects

• dcc-mcp-rpyc — RPyC bridge for remote DCC operations
• dcc-mcp-maya — Maya MCP server implementation

MCP Specification Roadmap

The library currently implements MCP 2025-03-26 (Streamable HTTP). The ecosystem has since released:

AI Agents: Do NOT implement draft features manually. Wait for dcc-mcp-core to expose them via McpHttpServer. Track progress at the GitHub repository.

Common Pitfalls

1. scan_and_load returns (List[SkillMetadata], List[str]) — always unpack: skills, skipped = scan_and_load(...)
2. Prefer public HostExecutionBridge / dispatcher wiring; use DeferredExecutor only when following docs/guide/dcc-thread-safety.md low-level guidance
3. Register ALL actions before server.start() — server reads from registry at startup only
4. Use IpcChannelAdapter + DccLinkFrame for IPC (v0.14+) — FramedChannel/connect_ipc were removed in #251
5. ToolDispatcher(registry) takes ONE arg — no validator= parameter
6. Action naming: {skill_name.replace('-','_')}__{script_stem} (double underscore)
7. SKILL.md name must match parent directory name (agentskills.io spec)
8. allowed-tools in SKILL.md is space-separated string, not a list (agentskills.io spec)
9. DccServerBase provides all skill/lifecycle/gateway/hot-reload methods — don't reimplement
10. MCP 2025-06-18 removes JSON-RPC batching — do not implement batch calls manually
11. MCP-Protocol-Version header is mandatory in 2025-06-18 — handled by McpHttpServer internally