Skip to main content

dcc-mcp-marmoset

DCC-MCP · MARMOSET

Typed DCC-MCP control for Marmoset Toolbag 4.03+ and 5.x. A pure-Python Toolbag plugin executes mset calls on mset.callbacks.onPeriodicUpdate; an external DccServerBase process owns MCP, discovery, jobs, and gateway registration.

Part of the DCC-MCP host matrix

dcc-mcp-marmoset — Marmoset Toolbag adapter and typed PBR scene/rendering skills for DCC-MCP.

It is one of 47 host adapters in the DCC-MCP catalog. Every adapter speaks the same MCP protocol and builds on the same core runtime contract; each one exposes the tools its own host needs on top of that.

This block is generated from the catalog entry in dcc-mcp-catalog.yml. Re-run the generator after changing the catalog.

Showcase

DCC-MCP Marmoset CC0 PBR lookdev

The adapter reconstructed Poly Haven's Dirty Football in a live Toolbag 5.02 session: FBX import, explicit Albedo/Normal/Roughness/Metalness/Occlusion assignment, exact-object framing, missing-reference validation, scene save, and 1920x1080 camera render. See the source and validation record.

Install

python -m pip install dcc-mcp-marmoset

In Toolbag, choose Edit > Plugins > Show User Plugin Folder and copy that folder path. Review a non-mutating install plan with the exact Toolbag binary, external interpreter, and copied plugin folder, then execute it:

dcc-mcp-marmoset install --json --dry-run `
  --dcc-path "C:\absolute\path\to\toolbag.exe" `
  --python "$((Get-Command python).Source)" `
  --plugin-dir "C:\path\shown\by\Toolbag"
dcc-mcp-marmoset install --json --yes `
  --dcc-path "C:\absolute\path\to\toolbag.exe" `
  --python "$((Get-Command python).Source)" `
  --plugin-dir "C:\path\shown\by\Toolbag"

Choose Edit > Plugins > Refresh, then launch DCC-MCP. The plugin starts one host-bound adapter process per Toolbag process; relaunching the plugin reuses that runtime. Toolbag requires one visible plugin window to keep callbacks alive, so the adapter uses one compact status window with the connection state, Toolbag version, loopback port, and stop control. Closing it stops the plugin. Run dcc-mcp-marmoset verify --json to prove the typed bridge ping; installed files alone are not readiness. See install.md for status, upgrade, receipt-driven uninstall, rollback, and troubleshooting.

Agent workflow

Shell-capable agents use the shared CLI:

dcc-mcp-cli list
dcc-mcp-cli search --query "inspect Toolbag scene" --dcc-type marmoset
dcc-mcp-cli load-skill marmoset-scene --dcc-type marmoset
dcc-mcp-cli search --query "inspect Toolbag scene" --dcc-type marmoset
dcc-mcp-cli describe <tool-slug>
dcc-mcp-cli call <tool-slug> --json '{"max_objects":500}'

Use the exact slugs returned by search. IDE-only clients may connect to the gateway MCP endpoint at http://127.0.0.1:9765/mcp.

Bundled Skills

  • marmoset-scene — inspect/import/save scenes, assign explicit PBR maps, control visibility, and render cameras.
  • marmoset-lookdev — inspect material slots and frame a hero object or the whole scene.
  • marmoset-diagnostics — inspect GPU/renderer/preferences, find missing textures, release unused resources, inspect tooltip database availability, and control tooltip display.

Typed tools

  • marmoset_scene__ping
  • marmoset_scene__inspect_scene
  • marmoset_scene__import_model
  • marmoset_scene__create_pbr_material
  • marmoset_scene__set_visibility
  • marmoset_scene__save_scene
  • marmoset_scene__render_camera
  • marmoset_lookdev__inspect_materials
  • marmoset_lookdev__frame_subject
  • marmoset_diagnostics__inspect_runtime
  • marmoset_diagnostics__validate_assets
  • marmoset_diagnostics__set_display_tooltips
  • marmoset_diagnostics__free_unused_resources

The adapter deliberately exposes no arbitrary Python execution. Its bridge is loopback-only, uses a per-launch random token, caps messages at 1 MiB, and expires queued requests before host mutation. Camera rendering is an async Core job backed by a monolithic Toolbag render call; check job/output state before retrying after a timeout.

Development

vx uv venv .venv --python 3.12
vx uv pip install --python .venv -e ".[dev]"
.venv\Scripts\python -m pytest
.venv\Scripts\python -m ruff check src tests tools
.venv\Scripts\python -m ruff format --check src tests tools
.venv\Scripts\python tools\lint_skills.py

The host API contract follows Marmoset's official Python scripting guide and Toolbag 5 Python API reference.

Metadata

Release files for dcc-mcp-marmoset 0.1.3

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for dcc-mcp-marmoset 0.1.3
File Size Uploaded
dcc_mcp_marmoset-0.1.3.tar.gz 2.4 MB Details

Built distribution (wheel)

Table of built distributions (wheels) for dcc-mcp-marmoset 0.1.3
File Interpreter ABI Platform
dcc_mcp_marmoset-0.1.3-py3-none-any.whl Python 3 none any Details

Total release size: 2.5 MB

Release files / dcc_mcp_marmoset-0.1.3.tar.gz

Download URL dcc_mcp_marmoset-0.1.3.tar.gz
Size 2.4 MB
Tags Source
SHA-256 checksum
How to use checksums
a8b7a883807434e15dcca0c0343812373225c573fad8a73e99ec2ba7b774d333
BLAKE2b-256 checksum
How to use checksums
7f3ff3e3e93f0c0295d10045873896f33420fbb574f7bb01f058738b5836040f
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Oct 9, 2026.

Transparency log

Release files / dcc_mcp_marmoset-0.1.3-py3-none-any.whl

Download URL dcc_mcp_marmoset-0.1.3-py3-none-any.whl
Size 47.0 kB
Tags Python 3
SHA-256 checksum
How to use checksums
cfac8b01ab4fb5d0c1026e0be9ddf3b443606675445661c02141c21419181693
BLAKE2b-256 checksum
How to use checksums
5aeac709b9c2d63a78ae4799d1492ef2455956da7d521e1bdf66194646f3d092
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
Yes
Uploaded via twine/7.0.0 CPython/3.13.14

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

PyPI Publish Attestation

PyPI verified that this artifact, at this checksum, originated from the publisher listed below.

Signed by GitHub Actions, verified by PyPI on Oct 9, 2026.

Transparency log

Release history Release notifications | RSS feed

This release

0.1.3 This release

2 release files

0.1.2

2 release files

0.1.1

2 release files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page