houdini-gsplat-relighting

houdini-gsplat-relighting is a skill for Claude Code from dcc-mcp/dcc-mcp-houdini. It costs 75 tokens per session (1,406 once invoked), scanned A, original, MIT.

A Houdini workflow for changing the lighting of Gaussian Splats, which are 3D scenes made from many small rendered points. It prepares splats, relights them in Solaris and Karma, and can rasterize them in Copernicus.

In plain words
What is it for?
Use it to inspect and validate splats, prepare them with SideFX Labs, relight them, and produce rasterized results. It is not for training or importing new splats.
Why use it?
It helps apply controlled lighting changes to an existing captured splat while checking its point data, camera information, image quality, and subject coverage.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter.

Good fit Use it to inspect and validate splats, prepare them with SideFX Labs, relight them, and produce rasterized results. It is not for training or importing new splats.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/dcc-mcp/dcc-mcp-houdini/houdini-gsplat-relighting
Install

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.

Any agent
npx skills add dcc-mcp/dcc-mcp-houdini --skill houdini-gsplat-relighting
Clone the repo
git clone --depth 1 https://github.com/dcc-mcp/dcc-mcp-houdini

Made for: Claude Code.

Wrote 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.

agentmods badge for houdini-gsplat-relighting

README.md
[![agentmods](https://agentmods.dev/badge/skills/dcc-mcp/dcc-mcp-houdini/houdini-gsplat-relighting/github.svg)](https://agentmods.dev/skills/dcc-mcp/dcc-mcp-houdini/houdini-gsplat-relighting)
Your own site
<a href="https://agentmods.dev/skills/dcc-mcp/dcc-mcp-houdini/houdini-gsplat-relighting"><img src="https://agentmods.dev/badge/skills/dcc-mcp/dcc-mcp-houdini/houdini-gsplat-relighting/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.

agentmods 80×15 button for houdini-gsplat-relighting

Your own site · 80×15
<a href="https://agentmods.dev/skills/dcc-mcp/dcc-mcp-houdini/houdini-gsplat-relighting"><img src="https://agentmods.dev/badge/skills/dcc-mcp/dcc-mcp-houdini/houdini-gsplat-relighting.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 75 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,406 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00075 $0.01406
Opus 5 $0.00037 $0.00703
Sonnet 5 $0.00015 $0.00281
Haiku 4.5 $0.00007 $0.00141

Measured 12d ago against content hash 6c15fa932869, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

houdini-gsplat-relighting 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.

The scan reads SKILL.md. This mod also ships 5 executable files (scripts/create_gsplat_copernicus_raster.py, scripts/create_gsplat_relight_lop.py, scripts/gsplat_relighting.py, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

src/dcc_mcp_houdini/skills/houdini-gsplat-relighting/SKILL.md · 112 lines

How it starts

The opening of the file, as written. The whole thing — 112 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Gaussian Splat relighting

Use the typed tools in this package for the cross-context handoff:

  1. Inspect the SOP output and confirm point attributes and provenance before changing it. The preflight recognizes Houdini-native GSplats and standard 3DGS PLY exports (f_dc*, rot*, scale*, opacity, and f_rest*). A public reconstruction showcase requires splats trained from at least three views of the same subject with solved camera poses. It also requires a typed camera-pose source and validation state when the capture uses a calibrated or estimated turntable. Estimated turntable poses remain blocked from public showcase use until optimizer or reprojection validation passes. It requires complete subject coverage and held-out evaluation over at least eight views with PSNR >= 25, SSIM >= 0.8, and LPIPS <= 0.2. These are publication gates, not guarantees that anatomy or identity is correct. Anatomy fidelity is a separate mandatory public-showcase gate: all evaluated regions from the fixed 12-region checklist must pass, silhouette IoU must be >= 0.90, normalized landmark error <= 0.03, thin-structure recall >= 0.85, and at least three novel views must be evaluated.
  2. Prepare it with Houdini Bake GSplats when normalization is required, then Labs Normals from GSplats and/or Delight GSplats. Keep spherical harmonics enabled when the source provides f_rest* coefficients.
  3. In Solaris, use Relight GSplats with USD lights, a render camera, shadows, shadow bias, and optional dome/HDRI lighting. Use houdini-parameters for version-specific Labs parameters that are not exposed by the setup tool.
  4. In Copernicus, import the prepared or relit SOP result and use Rasterize GSplats with camera metadata. The setup tool can append Sharpen, HSV, Gamma, and Premult nodes and set the H22 network resolution; use parameter skills for further interactive tuning.

The expected handoff attributes are P, Cd or albedo, N, orient, scale/pscale, opacity, and optional GS_SPH_R/G/B plus ao. The setup tools resolve Labs node type aliases at runtime because Labs asset namespaces vary between Houdini 22 builds.

Procedural meshes sampled into points are synthetic point clouds, not captured Gaussian Splat reconstructions. They may test relighting mechanics, but must not be presented as reconstruction evidence or public GSplat showcase input.

The fixed anatomy checklist covers the head capsule, compound eyes, antennae, mouthparts, thorax, abdomen, forewings, hindwings, forelegs, middle legs, and hind legs, plus tarsi/claws as an independent region across all six legs. anatomy_region_count may include additional documented regions, but anatomy_regions_passed must equal the total evaluated count. The bounded public inputs accept at most 64 regions and 64 novel views; normalized metrics must be finite values in [0, 1].

Read the full file on GitHub · 112 lines

Files

What ships with it

7 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.

Changes

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.

  1. 12d ago First seen · 112 lines · 75 tokens per session scan A 6c15fa932869

Subscribe to this mod's changes

houdini-gsplat-relighting is a skill published in the GitHub repository dcc-mcp/dcc-mcp-houdini (12 stars, last pushed 2d ago), licensed MIT. It adds 75 tokens to every session and 1,406 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.

Related

Other skills, from other repositories

maya-import-to-scene

Pipeline stage — structured asset import. Consume an AssetDescriptor produced by maya-asset-source and import the asset (FBX, OBJ, USD) into the current Maya scene via cmds.file(). Handles axis/unit conversion, MaterialMode, PlacementHint, and optional target collection grouping. Returns an ImportToSceneResult with…

dcc-mcp/dcc-mcp-maya · 78 tokens

usd-tools

Infrastructure skill — low-level OpenUSD scene inspection and validation: read layer stacks, traverse prims, validate USD schemas. Use when working directly with raw USD files (usda, usdc, usdz) or verifying USD compliance. Not for Maya-specific USD export — use maya-pipelineexportusd for that. Not for full DCC…

dcc-mcp/dcc-mcp-core · 83 tokens

maya-rigging

Authoring stage — character / prop rigging: joints, IK, skin clusters, deformers, blend shapes, control curves, skin weights, constraints, and optional rig framework detection. Use when constructing rigs. Not for keyframe animation (maya-animation), pose libraries (maya-pose-library), or final scene assembly…

dcc-mcp/dcc-mcp-maya · 78 tokens

maya-mesh-ops

Authoring stage — typed polygon construction and editing: loft, lathe, instance arrays, pivots, mirror, combine, separate, and cleanup. Use for creating or modifying polygon topology from explicit scene inputs. Not for primitive creation (use maya-primitives), construction-history or DG inspection (use…

dcc-mcp/dcc-mcp-maya · 90 tokens

maya-dynamics

Authoring stage - classic Maya dynamics primitives: rigid bodies, force fields, and field connections. Use for small simulation setup steps before baking with maya-animation. Not for keyframe editing (maya-animation), mesh modeling (maya-mesh-ops), or viewport output (maya-render).

dcc-mcp/dcc-mcp-maya · 63 tokens

maya-pose-library

Authoring stage — save / load / mirror character poses as JSON presets. Use for reusable pose libraries on rigged characters. Not for keyframe animation (maya-animation) or rigging setup (maya-rigging).

dcc-mcp/dcc-mcp-maya · 50 tokens