Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add winyunq/UnrealMotionGraphicsMCP --skill umg-material-mcpgit clone --depth 1 https://github.com/winyunq/UnrealMotionGraphicsMCPWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/winyunq/unrealmotiongraphicsmcp/umg-material-mcp)<a href="https://agentmods.dev/skills/winyunq/unrealmotiongraphicsmcp/umg-material-mcp"><img src="https://agentmods.dev/badge/skills/winyunq/unrealmotiongraphicsmcp/umg-material-mcp/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/winyunq/unrealmotiongraphicsmcp/umg-material-mcp"><img src="https://agentmods.dev/badge/skills/winyunq/unrealmotiongraphicsmcp/umg-material-mcp.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00084 | $0.00897 |
| Opus 5 | $0.00042 | $0.00449 |
| Sonnet 5 | $0.00017 | $0.00179 |
| Haiku 4.5 | $0.00008 | $0.00090 |
Grade A, and why
umg-material-mcp scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.
UMG Material MCP
Use the Material HLSL MCP path for material authoring. Do not mutate .uasset files directly. Keep Unreal Editor open with the UMG MCP bridge listening before runtime tool calls.
For concrete command examples and regression cases, read demo-cases.md.
Primary Workflow
-
Set or create the HLSL target with
hlsl_set_target.- New targets default to UI material behavior.
- For non-UI materials, pass type fields directly on
hlsl_set_target:domain,blend_mode,shading_model,two_sided. - Prefer this over
material_modify_type; that command is compatibility-only.
-
Read with
hlsl_get.- Treat the response as semantic state: HLSL text, parameter snapshot, and
output_contract. - Do not request or reconstruct full material graphs unless the HLSL protocol cannot express the task.
- Treat the response as semantic state: HLSL text, parameter snapshot, and
-
Write with
hlsl_set(hlsl?, parameters?, outputs?).hlsl_setis union-only: update existing entries and append missing entries.- It must not delete parameters or outputs.
- Use
parametersfor Custom node inputs such as{"name":"Intensity","kind":"Scalar"}. - Use
outputsfor root material outputs such as"Roughness","Metallic","Normal", or objects withname,target, andtype. - Additional outputs are UE5.8 Custom node
AdditionalOutputs; assign same-name variables in HLSL, for exampleRoughness = 0.35;.
-
Delete with
hlsl_delete(names, confirm_delete=true, kind?).- Omit
kindin the normal path. - If a name matches both parameter and output, expect failure with
requires_kind: true; retry withkind="parameter"orkind="output". - Do not use
hlsl_delete_parameterorhlsl_delete_outputin primary flows; they are compatibility aliases.
- Omit
-
Compile with
hlsl_compile.- Treat compile diagnostics as the feedback loop.
- Prefer small repair writes with
hlsl_set(hlsl=...)rather than resetting the target.
Protocol Rules
- The material is represented as one HLSL program on a single Custom node.
- The Custom node primary return is
float4. output_contractdescribes where rgb/a and any additional outputs are wired.- UI, PostProcess, LightFunction, and Unlit output rgb to
EmissiveColor; lit Surface output rgb usually maps toBaseColor. - Alpha maps only when semantically useful: UI/translucent to
Opacity, masked toOpacityMask. - Do not fall back to legacy
material_*graph editing unless the requested operation cannot be modeled by HLSL target, parameters, outputs, delete, and compile.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 60 lines · 84 tokens per session scan A c3bd1b0e6c8d
umg-material-mcp is a skill published in the GitHub repository winyunq/UnrealMotionGraphicsMCP (198 stars, last pushed 1mo ago), licensed MIT. It adds 84 tokens to every session and 897 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
develop-web-game
Use when Codex is building or iterating on a web game (HTML/JS) and needs a reliable development + testing loop: implement small changes, run a Playwright-based test script with short input bursts and intentional pauses, inspect screenshots/text, and review console errors with rendergametotext.
clay
AI 3D model generation agent. Generates text-to-3D and image-to-3D code (Python/JS/OpenSCAD) using Meshy, Tripo, Hunyuan3D, Rodin, Sloyd, and Stability APIs. Handles game pipeline integration, LOD, retopology, UV, and QC validation.
loot-agent-mcp
You are playing LOOT on Robinhood Chain (4663) through the loot-agent-mcp tools. The game in one line: hold tokens through risk to mint loot boxes, play to earn keys, open boxes with keys, claim your pro rata share of the token's fee pot in ETH. Pots grow when the token trades. Selling burns your game position for the…
blender-motion-state-inspection
Use this skill when inspecting Blender characters, rigs, poses, animation retargeting, ground contact, facing direction, or model-vs-motion alignment where screenshots alone are not enough.
develop-web-game
Build or iterate on HTML/JS web games with Playwright input bursts, gameplay screenshots, state inspection, and console checks.
game-redemption-codes
A Chinese-language skill for finding game redemption codes, also called gift codes or CDKs, with recent results searched on Bilibili.