inspect-vrchat-avatar-project

Inspect and document VRChat Avatar Unity projects with evidence-labeled audits.

45|Updated Jun 22, 2026
One-click install
npx skills add https://github.com/XiaoboooOvO/VRChatEditorSkill --skill inspect-vrchat-avatar-project-xiaoboooovo
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: inspect-vrchat-avatar-project
Source: https://github.com/XiaoboooOvO/VRChatEditorSkill
Command: npx skills add https://github.com/XiaoboooOvO/VRChatEditorSkill --skill inspect-vrchat-avatar-project-xiaoboooovo

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? VRChat Avatar Unity projects accumulate menus, parameters, Animator controllers, and third-party providers (FaceEmo, lilycalInventory, Modular Avatar) whose actual behavior is hard to trace. This Skill audits an avatar from the player-visible Expressions Menu down to exact model bindings, producing durable, evidence-labeled project documentation instead of guesswork. ## Core Features & Use Cases - Toolchain confirmation gate: Detects installed packages and provider components, then lets the user confirm tool roles before any audit or mutation. - Menu-to-model tracing: Follows each menu control through parameters, Animators, and providers to exact objects, materials, BlendShapes, PhysBones, and Contacts. - Evidence ladder: Labels every conclusion as STATIC_SOURCE, UNITY_RESOLVED, PROVIDER_PREVIEW, NDMF_BUILT, SDK_BUILD, CLIENT_RUNTIME, or UPLOAD_CONFIRMED, and marks unrun layers NOT_RUN or BUILD_REQUIRED. - Use Case: After importing a new clothing prefab, ask the Skill to trace each wardrobe toggle to its actual model objects, record the parameter delta, and refresh the audit documents. ## Quick Start Ask the agent to inspect this VRChat Avatar project, confirm the toolchain selection first, then audit the complete player menu and generate evidence-labeled documentation.

Frequently Asked Questions about inspect-vrchat-avatar-project

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I audit a VRChat Avatar Unity project?

Connect the project's Unity MCP instance, confirm the toolchain profile with the user, then trace the complete Expressions Menu tree to parameters, controllers, and model bindings. The Skill produces durable audit documents covering menus, parameters, providers, and evidence status.

How do I trace a VRChat menu toggle to the model object it controls?

Start from the full player menu path, follow the control value to its expression parameter, then to the owning provider, Animator state or clip, and the exact object, material, or BlendShape binding. Generated menus require an NDMF build check for final reachability.

Does this Skill work with FaceEmo, lilycalInventory, and Modular Avatar?

Yes, it ships dedicated playbooks for FaceEmo expressions, lilycalInventory wardrobes, and Modular Avatar, plus extended support for DressingTools, Gesture Manager, and Avatar Optimizer. A playbook activates only when the user selects that tool and the package is actually installed.

Can I use this Skill for VRChat World or Udon projects?

No. The workflow applies only to VRChat Avatar SDK projects. World scenes, Udon systems, and world runtime behavior are explicitly out of scope and should be routed to a dedicated World workflow.

What happens if Unity MCP is not connected?

Only static file inspection of serialized assets can proceed. Any conclusion requiring Editor-resolved state, previews, builds, or project modification is marked MCP_REQUIRED or BLOCKED, and the Skill will not substitute batch-mode Unity scripts for a real MCP connection.

Why does the Skill report NOT_RUN or BUILD_REQUIRED instead of a result?

Because source files cannot prove generated or runtime behavior. When a menu merge, parameter cost, or size claim needs a preview, NDMF build, SDK build, or runtime test that was not authorized or executed, the Skill reports the unrun layer explicitly rather than guessing.