Pick the right color management config (sRGB / Filmic / ACEScg / Standard) per output context before render, so colors travel correctly to the consumer.
Validate per-mesh and total-scene triangle counts against the pinned budget; flag worst offenders and topology hints without modeling.
Validate lighting intensity ranges, shadow softness, color temperature, and HDRI use against the scene brief before render.
Document GLB animation contract (skeletal vs morph, FPS, looping, root motion, blend transitions) with truthful caveats for web playback.
Verify chosen assets share visual style (realism, detail, color treatment) before scene build; flag style drift between assets early.
Set explicit triangle / texture / draw-call / file-size / animation budgets for GLB targeted at mobile-web before any export decision.
Validate PBR material sanity, texture budgets, and web/runtime compatibility per material before render or export.
Match user goals to BlendOps recipes and packs; flag scope gaps; recommend closest canonical workflow and skill chain.
Pick the right render resolution + aspect ratio for the consumer use case before render; avoid render-then-crop and aspect mismatch.
Turn a locked 3D brief plus chosen recipe into a measurable acceptance checklist with evidence hooks.
Detect and resolve MCP bridge conflicts (Path 1 Lab MCP + Path 2 ahujasid running together, port 9876 collisions, host mismatch).
Plan the React Three Fiber component shape (props, refs, Suspense, useGLTF cache, animation hooks) before delivering a GLB to a web team.
Plan headless blender --background export or render jobs when MCP chat control is a poor fit; cite upstream Blender CLI docs only.
Verify preview/render/GLB artifact truth by recording commands, inputs, output paths, file existence, validation notes, limitations, and conservative status labels.
Plan a wide environment / establishing-shot scene (locale, time of day, weather, focal subject, depth, atmosphere) before any modeling or render attempt.
Blender skills pack for non-Blender users. Use for scene planning, runtime readiness, render/export/GLB evidence, Claude Connect guidance, and plain-language handoff across…
Apply multi-category readiness gates with pass/warn/fail and evidence-bound verdicts.
Choose Path 1 Lab MCP, Path 2 ahujasid bridge, or CLI appendix from constraints; Connector hosts Path 1 only with Lab add-on inside Blender.
Establish official runtime prerequisites and setup boundaries before any Blender execution claim.
Verify brief completeness, runtime posture, skill coverage, and evidence habits before starting a pack workflow.
Plan the asset acquisition strategy (procedural, library, generative, photogrammetry) for a Blender scene without installing or executing.
Outline Path 1 hosting options — Anthropic Blender Connector vs manual MCP client — both need Blender Lab MCP add-on in Blender 5.1+.
Verify license compliance for downloaded 3D assets, textures, HDRIs, fonts before scene build or export.
Given current artifacts and intent, name the next BlendOps skill or workflow stage and preconditions to advance.
Prepare truthful GLB/web handoff output with explicit artifact status, assumptions, and caveats.
Generate the structured preview-report Markdown that glb-web-handoff and non-blender-user-response-writer consume after a render or export run.
Define explicit per-stage gates and acceptance checklists before any modeling, render, or export work — work only proceeds gate by gate.
Detect mid-recipe scope creep and push back constructively before the recipe drifts; require explicit user agreement before scope expansion.
Pick the right output file format (PNG / JPG / WebP / OpenEXR / TIFF) per use case based on bit depth, compression, transparency, and consumer support.
Define explicit "done" conditions for a Blender recipe before starting work, so the agent stops at the right point instead of drifting into endless polish.
Build detailed lighting and material direction with portability and caveat handling for Blender/web-3D outputs.
Break a complex / over-scoped Blender intent into 2-5 smaller recipes that each have a clean stop condition, brief, and handoff.
Convert a non-Blender-user's natural-language request into a structured 3D brief (subject, mood, deliverables, constraints, evidence expectations).
Validate camera framing, subject placement, headroom/lead-room, and visual hierarchy against the scene brief before render.
Plan an interior architectural visualization scene (room layout, furniture, materials, natural + artificial lighting) before any modeling or render attempt.
Plan an internal asset library structure (folders, naming, metadata, license tracking, version control) so future recipes find and reuse assets reliably.
Plan a multi-product grid scene (e-commerce, catalog, comparison) before any modeling or render attempt; consistent lighting, camera, and grid spacing.
Help users and agents choose the next safe BlendOps action across skills, docs, runtime readiness, evals, packaging, and Draft v0 release boundaries.
Provide a fallback strategy when an asset is unavailable, over-budget, license-blocked, or style-incompatible — without halting the recipe.
Generate a complete product-hero scene plan from non-Blender-user intent with explicit assumptions and gates.
Plan a character-portrait scene (subject, pose, expression, wardrobe, framing, lighting, background) before any modeling or render attempt.
Pick Cycles or Eevee (or Eevee Next / Workbench) for a render based on quality bar, time budget, and feature requirements (caustics, subsurface, etc.).
Claude Desktop-facing BlendOps router skill for planning, runtime readiness, evidence truth, and non-Blender-user outputs using bundled references.
Evaluate whether official runtime appears ready using evidence-bound checks and explicit readiness states.
Convert technical Blender workflow findings into plain-language user-facing responses with honest caveats.
Produce an evidence-bound runtime attempt report with truth labels and next steps; does not certify upstream stack success.
Provide detailed composition and camera planning for product hero outputs.