product-hero-scene-planner

name	product-hero-scene-planner
description	Generate a complete product-hero scene plan from non-Blender-user intent with explicit assumptions and gates.

product-hero-scene-planner

Purpose

Turn user intent into a high-quality, gate-ready scene plan before runtime execution.

Canonical role: scene-planning for product-hero workflows in the BlendOps skill system. Use this with the composition, camera, lighting, and material planners instead of creating duplicate generic scene-planning skills.

Quick start

• confirm this skill fits your goal
• provide required inputs first
• keep runtime claims evidence-bound
• follow suggested next-skill handoff

When to use

• product hero scenario requests
• pre-runtime plan generation
• requirements normalization for non-technical users

When not to use

• to claim generated artifacts
• to skip unresolved ambiguity

Trigger phrases

• “plan a product hero scene”
• “turn this into a Blender plan”
• “prepare cyberpunk shoe scene plan”

Prerequisites / readiness

• runtime status known (Ready/Partially Ready/Blocked/Unknown)
• output destination clarified
• no runtime execution required in this skill

Input schema

Required inputs

• user prompt
• output destination (web/app/game/render)

Optional inputs

• style/mood references
• product constraints
• performance constraints

Assumptions to confirm

• source model availability
• quality priority (realtime vs marketing)
• acceptable caveat level

Output schema

Primary output

• full scene plan schema

Secondary output

• clarification question set
• assumptions register

Evidence / caveat output

• unverified sections list
• runtime dependency caveats

Required laws

• ../../laws/official-runtime-only.md
• ../../laws/evidence-before-done.md
• ../../laws/non-blender-user-language.md
• ../../laws/no-arbitrary-python-interface.md

Official runtime boundary

Planning-only skill. Runtime execution and artifact claims are out of scope here.

Operating procedure

1. Restate user intent and desired outcome.
2. Identify target audience/use-case profile.
3. Extract hard constraints (style, brand, performance).
4. Generate clarification questions for missing critical data.
5. Build subject model (primary object and context elements).
6. Build composition plan (focal hierarchy, framing, negative space).
7. Build camera plan (angle, lens intent, depth strategy).
8. Build lighting plan (key/fill/rim + mood intent).
9. Build material plan (lookdev intent + portability caveats).
10. Build environment plan (background/support elements).
11. Define render/export intent and web handoff intent.
12. Attach validation-gate expectations and caveat policy.

Decision tree

• If destination is unknown → request destination first.
• If constraints conflict → produce option A/B with trade-offs.
• If runtime blocked → keep planning mode and mark artifacts Not Run.

Playbooks

• Playbook A: text-only planning path
• Playbook B: runtime-ready path with evidence gating
• Playbook C: blocked runtime path with caveat-first reporting

Mode handling

Text-only mode

• complete planning output
• artifact status remains Not Run

Runtime-ready mode

• include handoff steps for downstream runtime phase
• no success claim until evidence exists

Blocked runtime mode

• deliver complete plan + blocker-aware next actions

Validation checklist

• intent summary complete
• audience/use-case model included
• clarification questions provided when needed
• composition plan included
• camera plan included
• lighting plan included
• material plan included
• environment plan included
• render/export intent included
• web handoff intent included
• assumptions and caveats explicit
• no runtime output overclaims

Pass / Warn / Fail rubric

Category	Pass	Warn	Fail
Intent capture	Complete and coherent	Minor ambiguity remains	Core intent unclear
Plan completeness	All schema sections present	Some sections shallow	Critical sections missing
Assumption transparency	Explicit assumptions/caveats	Partial assumptions	Hidden assumptions
Evidence boundary	No overclaims	Slight ambiguity	Runtime/output success claimed without evidence

Failure handling

• Missing critical inputs: block and ask targeted questions.
• Conflicting constraints: provide prioritized options with caveats.
• Unknown runtime readiness: continue in text-only planning mode.

Troubleshooting

• Prompt too broad: split by objective and destination.
• Web + cinematic conflict: ask priority ranking.
• No source model: include intake risk and fallback plan.

Best practices

• preserve user language in summaries
• keep plan modular and checklist-ready
• surface unknowns early

Good examples

“Plan includes subject/composition/camera/lighting/material/environment plus explicit Not Run artifact status pending runtime phase.”

Bad examples

“Hero scene done.” (no structure, no caveats)

User-facing response template

• What you asked for
• Proposed plan
• What is not verified yet
• Next action

Anti-patterns

• skipping required laws or runtime boundary statements
• claiming runtime/artifact success without evidence
• using non-official runtime setup paths
• producing jargon-heavy final output without explanation

Cross-skill handoff

• Next: blender-composition-camera-planner
• Then: blender-lighting-material-planner
• Then: blender-scene-quality-checker

Non-goals

• runtime command execution
• artifact success claims

References

• https://www.blender.org/lab/mcp-server/
• https://claude.com/resources/tutorials/using-the-blender-connector-in-claude
• https://docs.blender.org/manual/en/latest/advanced/command_line/index.html