SkillAgentSearch skills...

scenario-maya-expert

Use when doing any Autodesk Maya task through Python or a bridge, headless or live: maya.cmds or OpenMaya 2.0 scripts, mayapy or maya.standalone batch jobs, a commandPort or MCP connection into a running Maya, "Maya 2027", validating or exporting scenes, or a maya.cmds script that fails after the 20…

Install / Use

npx skills add scenario-labs/skills --skill scenario-maya-expert

Installs into whichever agent you are using.

About this skill
πŸ“„

SKILL.md

Installable skill definition

Quality Score

86/100

Supported Platforms

Universal

Our assessment of scenario-maya-expert

scenario-maya-expert scores 86/100 on our quality scale, 2372nd of 4,705 Development & Engineering skills we index.

Its SKILL.md is 22 KB long, well organised into 8 sections and no code examples: a thorough specification that gives an agent plenty to work with.

With 1,048 GitHub stars, it is one of the more widely adopted skills in the catalogue.

Substance
30/30
Structure
13/20
Description
15/15
Adoption
13/20
Freshness
15/15

Maintenance, license and trust

  • The repository was last updated 35 days ago, so scenario-maya-expert is actively maintained.
  • It is released under the MIT license, a permissive license that allows use, modification and commercial use with attribution.
  • Its trust signals score 100/100, with no cautions. These come from repository metadata, not a code audit β€” read the skill file before letting an agent act on it.

Safety scan

No issues found

Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands.

Automated pattern scan on 2026-11-20. It catches known dangerous patterns, not every risk β€” read a skill before letting an agent act on it.

scenario-maya-expert compared with similar skills

All 4 of these similar skills score higher than scenario-maya-expert; compare them before choosing.

SkillScoreStarsUpdatedFormat
scenario-maya-expert (this skill)by scenario-labs861.0k35d agoSKILL.md
taste-skillby Leonxlnx100120.0ktodayCLAUDE.md
Agent-Reachby Panniantong100114.0ktodayCLAUDE.md
hyperframesby heygen-com10088.2ktodayCLAUDE.md
rufloby ruvnet10082.3ktodayCLAUDE.md

Frequently asked questions

How do I install scenario-maya-expert?
Run npx skills add scenario-labs/skills --skill scenario-maya-expert. The install tabs above show the steps for each supported agent.
Which AI agents does scenario-maya-expert work with?
It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
Is scenario-maya-expert safe to use?
Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. It is MIT-licensed and scores 100/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
Is scenario-maya-expert still maintained?
The repository was last updated 35 days ago, so scenario-maya-expert is actively maintained.

name: scenario-maya-expert description: 'Use when doing any Autodesk Maya task through Python or a bridge, headless or live: maya.cmds or OpenMaya 2.0 scripts, mayapy or maya.standalone batch jobs, a commandPort or MCP connection into a running Maya, "Maya 2027", validating or exporting scenes, or a maya.cmds script that fails after the 2025 to 2027 changes (PySide6, OpenPBR, renamed math nodes, removed flags). Also when a Maya model, rig, shot or render must be judged like a professional before it is called done, or a brief must be routed to the right Maya specialist.' license: MIT

Maya expert (router and agent protocol)

Expert Maya work is a loop, not a script: block big to small, and at every stage measure, render, look and fix before adding detail. This skill is the protocol every Maya task follows, the shared toolkit the other scenario-maya-* skills call, and the map to them. Target: Maya 2027 on macOS Apple Silicon, driven through Python. If a sibling skill named here is missing from your available skills, load it with skill_get (Scenario MCP) or ask the user to install it (https://github.com/scenario-labs/skills#install); unattended, proceed from tool schemas and flag the gap.

Status (2026-09-24): Maya 2027 is not installed yet. Every Maya call in scripts/ and references/ is not yet run in Maya. The pure-Python parts and the Maya-layer logic ran offline against fakes and pass. First action once Maya is installed: tests/code/maya-expert/run_all.sh, then read archive/tests/maya-expert/<stamp>/results/test_00_probe.json and correct anything marked [verify] here.

1. Execution channels

| Channel | Use for | Rules | | -------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Headless mayapy through scripts/mx_run.py | audits, validation, exports, Arnold review renders, batch work, anything that needs no viewport | python3 mx_run.py [--scene f.ma] [--plugins mtoa,fbx,abc,usd] job.py -- args. One child process per scene: a crash or hang costs one job. The last stdout line is MX_RESULT {json}; exit 0 ok, 1 job error, 4 timeout. Children get MAYA_DISABLE_ADP=1 and Maya's stderr at warning (--maya-log all to debug). Script nodes off. Never overwrite the source: --save-as a new version. | | Live GUI Maya through scripts/mx_bridge.py | playblasts, viewport captures, interactive tools, the user's open scene | Emmanuel starts mx_bridge_server.start() once in the Script Editor. Then Bridge().run(code): any length, multi-statement; every record carries result, stdout, stderr, maya_messages (Maya's warnings and errors) and the traceback, even on a timeout. Read them every call. One call = one undo chunk. Check cmds.file(q=True, modified=True) before replacing a scene. Never open modal UI. stop() when done. |

Do not launch a GUI Maya yourself [added]: it takes a license seat and cannot be driven without the bridge. One Arnold render at a time per machine [added].

mayapy: /Applications/Autodesk/maya2027/Maya.app/Contents/bin/mayapy (Maya 2027 Help). mx_run.find_mayapy() checks $MX_MAYAPY, $MAYA_LOCATION/bin/mayapy (on macOS MAYA_LOCATION ends in Maya.app/Contents), then the newest /Applications/Autodesk/maya20*/. Log cmds.about(version=True) at session start: Smart Bevel, MtoA and USD for Maya differ across 2027.x.

Why files for the bridge: a commandPort defaults to MEL, a 4096-character buffer (longer commands close the connection, longer results become an error) and returns the result of one statement only (Maya 2027 Help, commandPort; Python in Maya). mx_bridge sends one short line and exchanges code and results as files (GG_MayaMCP pre-installs a handler; Palmer uses two round trips). Security: the port is opened as 127.0.0.1:<port> [verify with lsof] and has no authentication; Maya's help warns an INET port runs any command, including system(), as the user. Personal machine only, close it after the session.

Capability matrix (from the docs; the probe test confirms or corrects each row):

| Action | Headless mayapy | GUI via bridge | Basis | | --------------------------------------------------------------------------------------------------- | --------------------------------------------------- | ---------------------------------------------------------------- | ------------------------------ | | cmds / OpenMaya edits, file IO, references, FBX, Alembic, USD export | yes | yes | Maya 2027 Help, Python in Maya | | Arnold stills (mx_review) | yes, CPU; watermark without an Arnold batch license | yes | Arnold batch docs | | Render -r hw2 viewport-style frames | documented, unverified on macOS | n/a | command-line render doc | | playblast, modelEditor, viewport grabs | no (needs a model panel) [verify] | yes, mx_review.playblast | scripting digest | | Tool contexts: Quad Draw, Multi-Cut, sculpt, Paint Skin Weights, Skin Tools, 3D paint, XGen brushes | no | interactive only (no stroke API): script the underlying commands | Maya 2027 Help | | scriptJob | no | yes, not during playback | Maya 2027 Help, scriptJobs | | maya.utils.executeInMainThreadWithResult | no, "not available in batch mode" | yes | Maya 2027 Help, threading | | maya.utils.executeDeferred | only when idle events run [verify] | yes | same | | MMessage callbacks | [verify] | yes | scripting digest | | commandPort | absent per Theodore 2014 [verify] | yes | MCP bridges note | | PySide6 windows, workspaceControl | no | yes | Qt6 migration docs | | Flow Retopology (cloud) | no: sign-in and network [added] | yes | version deltas | | confirmDialog, input(), pdb, file dialogs | never | never: they block Maya's main thread and the bridge | Maya 2027 Help, standard input |

2. The expert loop (every domain)

  1. Brief in numbers. Purpose (engine, film, still, animation), budgets (triangles, texture size and texel density, frame range and fps), style, references, deadline. When the brief is silent, take the domain skill's documented default and write it down.
  2. Stages from the domain skill, big to small. Never start with detail.
  3. Build one stage with small, idempotent scripts (get-or-create by name, tag what you create, selection-free).
  4. Gate the stage with measurable checks (mx_audit.verdict, mx_validate, the domain's numbers) AND a render you open with the image reader (mx_review.review sheet headless, or a playblast in the GUI). Experts turn the model constantly; the contact sheet is the agent's orbit. Judge it with the domain's references/critique.md.
  5. Fix before advancing. A problem found at blockout costs one script; found after texturing it costs the stage.
  6. Deliver with evidence: the saved version, audit and validation JSON, the sheet or playblast, and a list of what was not verified.

Never report a model, rig, shot or render as done without having looked at a render of it. Numbers first, then eyes: Fragapane's standard for a matching tool is the inverse of the expected matrix times the computed one equals identity, not "looks right" (Cult of Rig, [01:16:03]); dcc-mcp's showcase proves claims with deltas of 0.0 and color drift of at most 1/255.

3. Persona pipelines

| Brief | Skill chain | Deliverable | | -------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------- | | Hard-surface prop or game asset | scenario-maya-modeling, scenario-maya-retopology-uv, scenario-maya-lookdev, scenario-maya-pipeline-scripting (FBX) | low poly, UVs, PBR set, FBX re-imported and compared | | Finish an AI-generated mesh | scenario-3d, mx_audit on the raw mesh, scenario-maya-retopology-uv, scenario-maya-lookdev, scenario-maya-pipeline-scripting | clean, UV'd asset at a stated budget, baked from the generated high | | Character for keyframe animation | scenario-maya-modeling, scenario-maya-retopology-uv, scenario-maya-rigging, scenario-maya-deformation, scenario-maya-a

Truncated for display β€” read the full file on GitHub.

Related Skills

View on GitHub
GitHub Stars1.0k
CategoryDevelopment
Updated1mo ago
Forks129

Languages

Python

Trust signals

100/100

From repository metadata: license, adoption, age and documentation. Not a code audit β€” see the Safety scan above for what the skill file itself contains.

No cautions
scenario-maya-expert β€” Universal Skill: Install & Safety Check | SkillAgent