blendops 0.0.0-stage → 1.1.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CHANGELOG.md +74 -0
- package/LICENSE +21 -0
- package/README.md +208 -2
- package/bin/blendops.mjs +4 -0
- package/bundles/claude-desktop-manual/BLENDOPS_SINGLE_FILE.md +78 -0
- package/bundles/claude-desktop-manual/IMPORT_STEPS.md +43 -0
- package/bundles/claude-desktop-manual/INSTALL_REPORT_TEMPLATE.md +13 -0
- package/bundles/claude-desktop-manual/README.md +70 -0
- package/bundles/claude-desktop-manual/SKILL.md +62 -0
- package/bundles/claude-desktop-manual/manifest.json +278 -0
- package/bundles/claude-desktop-manual/references/evidence-rules.md +13 -0
- package/bundles/claude-desktop-manual/references/install-boundary.md +13 -0
- package/bundles/claude-desktop-manual/references/laws/evidence-before-done.md +14 -0
- package/bundles/claude-desktop-manual/references/laws/no-arbitrary-python-interface.md +12 -0
- package/bundles/claude-desktop-manual/references/laws/non-blender-user-language.md +13 -0
- package/bundles/claude-desktop-manual/references/laws/official-runtime-only.md +20 -0
- package/bundles/claude-desktop-manual/references/packs/product-hero-v0.md +30 -0
- package/bundles/claude-desktop-manual/references/runtime-stacks.md +13 -0
- package/bundles/claude-desktop-manual/references/skill-map.md +36 -0
- package/bundles/claude-desktop-manual/references/skills/acceptance-criteria-builder.md +69 -0
- package/bundles/claude-desktop-manual/references/skills/asset-discovery.md +73 -0
- package/bundles/claude-desktop-manual/references/skills/asset-fallback-strategy.md +66 -0
- package/bundles/claude-desktop-manual/references/skills/asset-library-plan.md +69 -0
- package/bundles/claude-desktop-manual/references/skills/asset-license-checker.md +69 -0
- package/bundles/claude-desktop-manual/references/skills/asset-style-check.md +69 -0
- package/bundles/claude-desktop-manual/references/skills/blender-brainstorming.md +62 -0
- package/bundles/claude-desktop-manual/references/skills/blender-recipe-decomposer.md +66 -0
- package/bundles/claude-desktop-manual/references/skills/blender-troubleshooting.md +78 -0
- package/bundles/claude-desktop-manual/references/skills/blendops-help.md +82 -0
- package/bundles/claude-desktop-manual/references/skills/bridge-conflicts.md +68 -0
- package/bundles/claude-desktop-manual/references/skills/camera-composition.md +65 -0
- package/bundles/claude-desktop-manual/references/skills/character-portrait.md +68 -0
- package/bundles/claude-desktop-manual/references/skills/cli-appendix-job-outline.md +62 -0
- package/bundles/claude-desktop-manual/references/skills/color-management-decision.md +68 -0
- package/bundles/claude-desktop-manual/references/skills/composition-quality-checker.md +71 -0
- package/bundles/claude-desktop-manual/references/skills/cycles-vs-eevee-decision.md +65 -0
- package/bundles/claude-desktop-manual/references/skills/done-criteria.md +70 -0
- package/bundles/claude-desktop-manual/references/skills/environment-shot.md +65 -0
- package/bundles/claude-desktop-manual/references/skills/glb-animation-handoff.md +69 -0
- package/bundles/claude-desktop-manual/references/skills/glb-web-handoff.md +68 -0
- package/bundles/claude-desktop-manual/references/skills/intent-to-3d-brief-writer.md +68 -0
- package/bundles/claude-desktop-manual/references/skills/interior-scene.md +73 -0
- package/bundles/claude-desktop-manual/references/skills/lighting-material-plan.md +65 -0
- package/bundles/claude-desktop-manual/references/skills/lighting-quality-checker.md +70 -0
- package/bundles/claude-desktop-manual/references/skills/material-quality-checker.md +71 -0
- package/bundles/claude-desktop-manual/references/skills/mobile-glb-budget.md +73 -0
- package/bundles/claude-desktop-manual/references/skills/output-format-decision.md +63 -0
- package/bundles/claude-desktop-manual/references/skills/pack-prerequisite-checker.md +74 -0
- package/bundles/claude-desktop-manual/references/skills/path-one-host-outline.md +66 -0
- package/bundles/claude-desktop-manual/references/skills/plain-language-response.md +70 -0
- package/bundles/claude-desktop-manual/references/skills/polycount-budget-checker.md +71 -0
- package/bundles/claude-desktop-manual/references/skills/pre-handoff-verification.md +76 -0
- package/bundles/claude-desktop-manual/references/skills/preview-report.md +62 -0
- package/bundles/claude-desktop-manual/references/skills/product-grid-scene-planner.md +72 -0
- package/bundles/claude-desktop-manual/references/skills/product-hero-scene-planner.md +68 -0
- package/bundles/claude-desktop-manual/references/skills/r3f-component-plan.md +66 -0
- package/bundles/claude-desktop-manual/references/skills/recipe-fit-assessor.md +75 -0
- package/bundles/claude-desktop-manual/references/skills/render-export-evidence.md +70 -0
- package/bundles/claude-desktop-manual/references/skills/resolution-aspect-decision.md +65 -0
- package/bundles/claude-desktop-manual/references/skills/runtime-attempt-report.md +67 -0
- package/bundles/claude-desktop-manual/references/skills/runtime-path-picker.md +67 -0
- package/bundles/claude-desktop-manual/references/skills/runtime-readiness.md +63 -0
- package/bundles/claude-desktop-manual/references/skills/runtime-setup.md +63 -0
- package/bundles/claude-desktop-manual/references/skills/scene-readiness.md +65 -0
- package/bundles/claude-desktop-manual/references/skills/scope-guard.md +62 -0
- package/bundles/claude-desktop-manual/references/skills/stage-gates.md +91 -0
- package/bundles/claude-desktop-manual/references/skills/workflow-stage-router.md +67 -0
- package/bundles/generic-project-local/BLENDOPS.md +71 -0
- package/bundles/generic-project-local/INSTALL_REPORT_TEMPLATE.md +52 -0
- package/bundles/generic-project-local/README.md +73 -0
- package/bundles/generic-project-local/ROLLBACK.md +66 -0
- package/bundles/generic-project-local/manifest.json +66 -0
- package/bundles/skill-package/README.md +54 -0
- package/bundles/skill-package/blendops/LICENSE.txt +21 -0
- package/bundles/skill-package/blendops/SKILL.md +50 -0
- package/bundles/skill-package/blendops/agents/README.md +61 -0
- package/bundles/skill-package/blendops/agents/openai.yaml +4 -0
- package/bundles/skill-package/blendops/references/evidence-rules.md +17 -0
- package/bundles/skill-package/blendops/references/install-boundary.md +13 -0
- package/bundles/skill-package/blendops/references/law-evidence-before-done.md +12 -0
- package/bundles/skill-package/blendops/references/law-no-arbitrary-python-interface.md +7 -0
- package/bundles/skill-package/blendops/references/law-non-blender-user-language.md +8 -0
- package/bundles/skill-package/blendops/references/law-official-runtime-only.md +9 -0
- package/bundles/skill-package/blendops/references/pack-product-hero-v0.md +21 -0
- package/bundles/skill-package/blendops/references/runtime-stacks.md +13 -0
- package/bundles/skill-package/blendops/references/skill-acceptance-criteria-builder.md +51 -0
- package/bundles/skill-package/blendops/references/skill-asset-discovery.md +54 -0
- package/bundles/skill-package/blendops/references/skill-asset-fallback-strategy.md +48 -0
- package/bundles/skill-package/blendops/references/skill-asset-library-plan.md +51 -0
- package/bundles/skill-package/blendops/references/skill-asset-license-checker.md +51 -0
- package/bundles/skill-package/blendops/references/skill-asset-style-check.md +52 -0
- package/bundles/skill-package/blendops/references/skill-blender-brainstorming.md +43 -0
- package/bundles/skill-package/blendops/references/skill-blender-recipe-decomposer.md +48 -0
- package/bundles/skill-package/blendops/references/skill-blender-troubleshooting.md +59 -0
- package/bundles/skill-package/blendops/references/skill-blendops-help.md +62 -0
- package/bundles/skill-package/blendops/references/skill-bridge-conflicts.md +48 -0
- package/bundles/skill-package/blendops/references/skill-camera-composition.md +48 -0
- package/bundles/skill-package/blendops/references/skill-character-portrait.md +50 -0
- package/bundles/skill-package/blendops/references/skill-cli-appendix-job-outline.md +44 -0
- package/bundles/skill-package/blendops/references/skill-color-management-decision.md +50 -0
- package/bundles/skill-package/blendops/references/skill-composition-quality-checker.md +53 -0
- package/bundles/skill-package/blendops/references/skill-cycles-vs-eevee-decision.md +47 -0
- package/bundles/skill-package/blendops/references/skill-done-criteria.md +53 -0
- package/bundles/skill-package/blendops/references/skill-environment-shot.md +47 -0
- package/bundles/skill-package/blendops/references/skill-glb-animation-handoff.md +51 -0
- package/bundles/skill-package/blendops/references/skill-glb-web-handoff.md +51 -0
- package/bundles/skill-package/blendops/references/skill-intent-to-3d-brief-writer.md +50 -0
- package/bundles/skill-package/blendops/references/skill-interior-scene.md +54 -0
- package/bundles/skill-package/blendops/references/skill-lighting-material-plan.md +48 -0
- package/bundles/skill-package/blendops/references/skill-lighting-quality-checker.md +52 -0
- package/bundles/skill-package/blendops/references/skill-map.md +32 -0
- package/bundles/skill-package/blendops/references/skill-material-quality-checker.md +53 -0
- package/bundles/skill-package/blendops/references/skill-mobile-glb-budget.md +55 -0
- package/bundles/skill-package/blendops/references/skill-output-format-decision.md +46 -0
- package/bundles/skill-package/blendops/references/skill-pack-prerequisite-checker.md +56 -0
- package/bundles/skill-package/blendops/references/skill-path-one-host-outline.md +48 -0
- package/bundles/skill-package/blendops/references/skill-plain-language-response.md +53 -0
- package/bundles/skill-package/blendops/references/skill-polycount-budget-checker.md +53 -0
- package/bundles/skill-package/blendops/references/skill-pre-handoff-verification.md +57 -0
- package/bundles/skill-package/blendops/references/skill-preview-report.md +45 -0
- package/bundles/skill-package/blendops/references/skill-product-grid-scene-planner.md +54 -0
- package/bundles/skill-package/blendops/references/skill-product-hero-scene-planner.md +51 -0
- package/bundles/skill-package/blendops/references/skill-r3f-component-plan.md +48 -0
- package/bundles/skill-package/blendops/references/skill-recipe-fit-assessor.md +57 -0
- package/bundles/skill-package/blendops/references/skill-render-export-evidence.md +51 -0
- package/bundles/skill-package/blendops/references/skill-resolution-aspect-decision.md +48 -0
- package/bundles/skill-package/blendops/references/skill-runtime-attempt-report.md +49 -0
- package/bundles/skill-package/blendops/references/skill-runtime-path-picker.md +49 -0
- package/bundles/skill-package/blendops/references/skill-runtime-readiness.md +45 -0
- package/bundles/skill-package/blendops/references/skill-runtime-setup.md +45 -0
- package/bundles/skill-package/blendops/references/skill-scene-readiness.md +48 -0
- package/bundles/skill-package/blendops/references/skill-scope-guard.md +44 -0
- package/bundles/skill-package/blendops/references/skill-stage-gates.md +73 -0
- package/bundles/skill-package/blendops/references/skill-workflow-stage-router.md +49 -0
- package/docs/README.md +276 -0
- package/docs/examples/glb-web-handoff/README.md +174 -0
- package/docs/examples/glb-web-handoff/r3f-glb-viewer.tsx +320 -0
- package/docs/examples/glb-web-handoff/threejs-glb-viewer.mjs +709 -0
- package/docs/knowledge-and-recipe-catalog.md +236 -0
- package/docs/recipes/README.md +27 -0
- package/docs/recipes/cyberpunk-shoe-hero.md +97 -0
- package/docs/workflows/README.md +14 -0
- package/docs/workflows/full-non-blender-user-workflow.md +210 -0
- package/docs/workflows/product-hero-workflow.md +92 -0
- package/laws/README.md +32 -0
- package/laws/evidence-before-done.md +37 -0
- package/laws/no-arbitrary-python-interface.md +32 -0
- package/laws/non-blender-user-language.md +32 -0
- package/laws/official-runtime-only.md +78 -0
- package/package.json +103 -4
- package/packs/README.md +11 -0
- package/packs/product-hero-v0/PACK.md +114 -0
- package/registry/catalog-index.json +21344 -0
- package/registry/compatibility-matrix.json +533 -0
- package/registry/contract-context-fields.json +391 -0
- package/registry/knowledge-catalog.json +4648 -0
- package/registry/knowledge-sources.json +438 -0
- package/registry/recipe-catalog.json +2565 -0
- package/registry/runtime-capability-policy.json +525 -0
- package/registry/runtime-operations.json +4468 -0
- package/registry/skills.json +10231 -0
- package/registry/workflow-graph.json +7390 -0
- package/schemas/adapter-compatibility-matrix.schema.json +166 -0
- package/schemas/contract-context-fields.schema.json +147 -0
- package/schemas/execution-plan.schema.json +325 -0
- package/schemas/glb-web-handoff-manifest.schema.json +744 -0
- package/schemas/knowledge-record.schema.json +460 -0
- package/schemas/knowledge-source-registry.schema.json +268 -0
- package/schemas/operation-execution-ledger.schema.json +348 -0
- package/schemas/recipe-record.schema.json +469 -0
- package/schemas/runtime-capability-policy.schema.json +184 -0
- package/schemas/runtime-evidence-packet.schema.json +211 -0
- package/schemas/runtime-operations.schema.json +535 -0
- package/schemas/runtime-qualification-report.schema.json +540 -0
- package/schemas/runtime-session.schema.json +350 -0
- package/schemas/scene-spec.schema.json +624 -0
- package/schemas/skill-contract.schema.json +402 -0
- package/schemas/skill-eval.schema.json +509 -0
- package/scripts/build-knowledge-catalog.mjs +2376 -0
- package/scripts/compile-scene-plan.mjs +1945 -0
- package/scripts/export-claude-skills.mjs +634 -0
- package/scripts/generate-glb-handoff-report.mjs +2376 -0
- package/scripts/inspect-glb.mjs +1627 -0
- package/scripts/install-skills.mjs +5443 -0
- package/scripts/runtime-orchestrator.mjs +7713 -0
- package/scripts/validate-capability-policy.mjs +531 -0
- package/scripts/validate-evidence-packet.mjs +256 -0
- package/skills/README.md +315 -0
- package/skills/_template/SKILL.md +296 -0
- package/skills/acceptance-criteria-builder/EVAL.md +25 -0
- package/skills/acceptance-criteria-builder/SKILL.md +257 -0
- package/skills/acceptance-criteria-builder/contract.json +205 -0
- package/skills/acceptance-criteria-builder/references/brief-to-criteria-mapper.md +22 -0
- package/skills/acceptance-criteria-builder/references/checklist-patterns.md +22 -0
- package/skills/acceptance-criteria-builder/references/evidence-hook-rules.md +19 -0
- package/skills/asset-discovery/EVAL.md +81 -0
- package/skills/asset-discovery/SKILL.md +252 -0
- package/skills/asset-discovery/contract.json +291 -0
- package/skills/asset-discovery/references/asset-strategy-decision-tree.md +40 -0
- package/skills/asset-discovery/references/budget-reconciliation-patterns.md +54 -0
- package/skills/asset-discovery/references/license-confidence-table.md +41 -0
- package/skills/asset-fallback-strategy/EVAL.md +34 -0
- package/skills/asset-fallback-strategy/SKILL.md +248 -0
- package/skills/asset-fallback-strategy/contract.json +190 -0
- package/skills/asset-fallback-strategy/references/drop-with-caveat-rules.md +36 -0
- package/skills/asset-fallback-strategy/references/fallback-option-types.md +55 -0
- package/skills/asset-fallback-strategy/references/scoring-rubric.md +61 -0
- package/skills/asset-library-plan/EVAL.md +28 -0
- package/skills/asset-library-plan/SKILL.md +263 -0
- package/skills/asset-library-plan/contract.json +177 -0
- package/skills/asset-library-plan/references/folder-structure-rules.md +63 -0
- package/skills/asset-library-plan/references/metadata-schema.md +62 -0
- package/skills/asset-library-plan/references/naming-rules.md +51 -0
- package/skills/asset-license-checker/EVAL.md +33 -0
- package/skills/asset-license-checker/SKILL.md +245 -0
- package/skills/asset-license-checker/contract.json +172 -0
- package/skills/asset-license-checker/references/attribution-rendering-patterns.md +48 -0
- package/skills/asset-license-checker/references/common-license-mapping.md +39 -0
- package/skills/asset-license-checker/references/license-source-rules.md +38 -0
- package/skills/asset-style-check/EVAL.md +26 -0
- package/skills/asset-style-check/SKILL.md +238 -0
- package/skills/asset-style-check/contract.json +251 -0
- package/skills/asset-style-check/references/drift-classification.md +40 -0
- package/skills/asset-style-check/references/intentional-mixed-style.md +39 -0
- package/skills/asset-style-check/references/style-axes.md +46 -0
- package/skills/blender-brainstorming/EVAL.md +79 -0
- package/skills/blender-brainstorming/SKILL.md +228 -0
- package/skills/blender-brainstorming/contract.json +115 -0
- package/skills/blender-brainstorming/references/alternative-scoping-templates.md +43 -0
- package/skills/blender-brainstorming/references/clarification-patterns.md +29 -0
- package/skills/blender-brainstorming/references/intent-summary-style-guide.md +25 -0
- package/skills/blender-recipe-decomposer/EVAL.md +35 -0
- package/skills/blender-recipe-decomposer/SKILL.md +258 -0
- package/skills/blender-recipe-decomposer/contract.json +152 -0
- package/skills/blender-recipe-decomposer/references/dependency-patterns.md +53 -0
- package/skills/blender-recipe-decomposer/references/recipe-grouping-rules.md +58 -0
- package/skills/blender-recipe-decomposer/references/recipe-naming-rules.md +44 -0
- package/skills/blender-troubleshooting/EVAL.md +78 -0
- package/skills/blender-troubleshooting/SKILL.md +263 -0
- package/skills/blender-troubleshooting/contract.json +352 -0
- package/skills/blender-troubleshooting/references/common-symptom-playbooks.md +51 -0
- package/skills/blender-troubleshooting/references/divergence-routing-table.md +23 -0
- package/skills/blender-troubleshooting/references/four-phase-template.md +51 -0
- package/skills/blendops-help/EVAL.md +67 -0
- package/skills/blendops-help/SKILL.md +312 -0
- package/skills/blendops-help/contract.json +229 -0
- package/skills/blendops-help/references/lifecycle-phases.md +91 -0
- package/skills/blendops-help/references/quick-start-prompts.md +65 -0
- package/skills/blendops-help/references/routing-decision-tree.md +57 -0
- package/skills/bridge-conflicts/EVAL.md +83 -0
- package/skills/bridge-conflicts/SKILL.md +246 -0
- package/skills/bridge-conflicts/contract.json +145 -0
- package/skills/bridge-conflicts/references/bridge-state-matrix-template.md +82 -0
- package/skills/bridge-conflicts/references/conflict-playbooks.md +99 -0
- package/skills/bridge-conflicts/references/do-not-reminders.md +42 -0
- package/skills/camera-composition/EVAL.md +32 -0
- package/skills/camera-composition/SKILL.md +196 -0
- package/skills/camera-composition/contract.json +180 -0
- package/skills/camera-composition/references/camera-presets.md +30 -0
- package/skills/camera-composition/references/camera-rubric.md +24 -0
- package/skills/camera-composition/references/hero-composition-patterns.md +24 -0
- package/skills/camera-composition/references/web-layout-negative-space.md +25 -0
- package/skills/character-portrait/EVAL.md +42 -0
- package/skills/character-portrait/SKILL.md +245 -0
- package/skills/character-portrait/contract.json +205 -0
- package/skills/character-portrait/references/portrait-framing-rules.md +39 -0
- package/skills/character-portrait/references/portrait-lighting-patterns.md +50 -0
- package/skills/character-portrait/references/style-consistency-rules.md +41 -0
- package/skills/cli-appendix-job-outline/EVAL.md +26 -0
- package/skills/cli-appendix-job-outline/SKILL.md +252 -0
- package/skills/cli-appendix-job-outline/contract.json +147 -0
- package/skills/cli-appendix-job-outline/references/flag-order-reminders.md +16 -0
- package/skills/cli-appendix-job-outline/references/job-type-matrix.md +13 -0
- package/skills/cli-appendix-job-outline/references/stateless-limitations.md +12 -0
- package/skills/color-management-decision/EVAL.md +28 -0
- package/skills/color-management-decision/SKILL.md +254 -0
- package/skills/color-management-decision/contract.json +162 -0
- package/skills/color-management-decision/references/look-comparison.md +44 -0
- package/skills/color-management-decision/references/per-channel-rules.md +49 -0
- package/skills/color-management-decision/references/pipeline-mapping.md +47 -0
- package/skills/composition-quality-checker/EVAL.md +38 -0
- package/skills/composition-quality-checker/SKILL.md +260 -0
- package/skills/composition-quality-checker/contract.json +221 -0
- package/skills/composition-quality-checker/references/aspect-ratio-rules.md +37 -0
- package/skills/composition-quality-checker/references/composition-rules.md +58 -0
- package/skills/composition-quality-checker/references/headroom-leadroom-rules.md +46 -0
- package/skills/cycles-vs-eevee-decision/EVAL.md +26 -0
- package/skills/cycles-vs-eevee-decision/SKILL.md +249 -0
- package/skills/cycles-vs-eevee-decision/contract.json +142 -0
- package/skills/cycles-vs-eevee-decision/references/engine-feature-matrix.md +39 -0
- package/skills/cycles-vs-eevee-decision/references/gpu-device-rules.md +51 -0
- package/skills/cycles-vs-eevee-decision/references/samples-by-quality.md +53 -0
- package/skills/done-criteria/EVAL.md +33 -0
- package/skills/done-criteria/SKILL.md +225 -0
- package/skills/done-criteria/contract.json +199 -0
- package/skills/done-criteria/references/done-vs-perfect.md +56 -0
- package/skills/done-criteria/references/out-of-scope-patterns.md +50 -0
- package/skills/done-criteria/references/stop-condition-templates.md +39 -0
- package/skills/environment-shot/EVAL.md +44 -0
- package/skills/environment-shot/SKILL.md +237 -0
- package/skills/environment-shot/contract.json +162 -0
- package/skills/environment-shot/references/depth-layering-rules.md +49 -0
- package/skills/environment-shot/references/time-and-weather-rules.md +35 -0
- package/skills/environment-shot/references/wide-shot-camera-rules.md +51 -0
- package/skills/glb-animation-handoff/EVAL.md +39 -0
- package/skills/glb-animation-handoff/SKILL.md +271 -0
- package/skills/glb-animation-handoff/contract.json +247 -0
- package/skills/glb-animation-handoff/references/animation-types.md +36 -0
- package/skills/glb-animation-handoff/references/transition-rules.md +55 -0
- package/skills/glb-animation-handoff/references/web-stack-caveats.md +40 -0
- package/skills/glb-web-handoff/EVAL.md +33 -0
- package/skills/glb-web-handoff/SKILL.md +212 -0
- package/skills/glb-web-handoff/contract.json +171 -0
- package/skills/glb-web-handoff/references/artifact-status-language.md +26 -0
- package/skills/glb-web-handoff/references/glb-handoff-checklist.md +30 -0
- package/skills/glb-web-handoff/references/performance-caveats.md +26 -0
- package/skills/glb-web-handoff/references/threejs-r3f-handoff.md +28 -0
- package/skills/intent-to-3d-brief-writer/EVAL.md +82 -0
- package/skills/intent-to-3d-brief-writer/SKILL.md +238 -0
- package/skills/intent-to-3d-brief-writer/contract.json +195 -0
- package/skills/intent-to-3d-brief-writer/references/acceptance-criteria-patterns.md +46 -0
- package/skills/intent-to-3d-brief-writer/references/eight-slot-template.md +123 -0
- package/skills/intent-to-3d-brief-writer/references/mixed-audience-handling.md +39 -0
- package/skills/interior-scene/EVAL.md +44 -0
- package/skills/interior-scene/SKILL.md +261 -0
- package/skills/interior-scene/contract.json +277 -0
- package/skills/interior-scene/references/furniture-layout-rules.md +51 -0
- package/skills/interior-scene/references/interior-architecture-rules.md +47 -0
- package/skills/interior-scene/references/interior-camera-rules.md +47 -0
- package/skills/lighting-material-plan/EVAL.md +32 -0
- package/skills/lighting-material-plan/SKILL.md +194 -0
- package/skills/lighting-material-plan/contract.json +213 -0
- package/skills/lighting-material-plan/references/cyberpunk-lookdev.md +22 -0
- package/skills/lighting-material-plan/references/lighting-patterns.md +30 -0
- package/skills/lighting-material-plan/references/material-portability.md +25 -0
- package/skills/lighting-material-plan/references/reflective-floor-pattern.md +22 -0
- package/skills/lighting-material-plan/references/web-material-simplification.md +23 -0
- package/skills/lighting-quality-checker/EVAL.md +41 -0
- package/skills/lighting-quality-checker/SKILL.md +247 -0
- package/skills/lighting-quality-checker/contract.json +187 -0
- package/skills/lighting-quality-checker/references/hdri-balance-rules.md +48 -0
- package/skills/lighting-quality-checker/references/intensity-ranges.md +73 -0
- package/skills/lighting-quality-checker/references/shadow-softness-rules.md +38 -0
- package/skills/material-quality-checker/EVAL.md +40 -0
- package/skills/material-quality-checker/SKILL.md +261 -0
- package/skills/material-quality-checker/contract.json +244 -0
- package/skills/material-quality-checker/references/consumer-stack-rules.md +62 -0
- package/skills/material-quality-checker/references/pbr-sanity-rules.md +78 -0
- package/skills/material-quality-checker/references/texture-budget-rules.md +48 -0
- package/skills/mobile-glb-budget/EVAL.md +40 -0
- package/skills/mobile-glb-budget/SKILL.md +264 -0
- package/skills/mobile-glb-budget/contract.json +309 -0
- package/skills/mobile-glb-budget/references/budget-tiers.md +30 -0
- package/skills/mobile-glb-budget/references/degradation-plan.md +34 -0
- package/skills/mobile-glb-budget/references/violation-rules.md +43 -0
- package/skills/output-format-decision/EVAL.md +27 -0
- package/skills/output-format-decision/SKILL.md +222 -0
- package/skills/output-format-decision/contract.json +179 -0
- package/skills/output-format-decision/references/bit-depth-rules.md +56 -0
- package/skills/output-format-decision/references/compression-tradeoffs.md +62 -0
- package/skills/output-format-decision/references/format-decision-matrix.md +34 -0
- package/skills/pack-prerequisite-checker/EVAL.md +26 -0
- package/skills/pack-prerequisite-checker/SKILL.md +272 -0
- package/skills/pack-prerequisite-checker/contract.json +214 -0
- package/skills/pack-prerequisite-checker/references/blocked-start-playbook.md +21 -0
- package/skills/pack-prerequisite-checker/references/pack-readiness-template.md +38 -0
- package/skills/pack-prerequisite-checker/references/prerequisite-matrix.md +20 -0
- package/skills/path-one-host-outline/EVAL.md +26 -0
- package/skills/path-one-host-outline/SKILL.md +251 -0
- package/skills/path-one-host-outline/contract.json +167 -0
- package/skills/path-one-host-outline/references/connector-vs-manual-matrix.md +21 -0
- package/skills/path-one-host-outline/references/host-checklist-template.md +35 -0
- package/skills/path-one-host-outline/references/lab-addon-reminders.md +12 -0
- package/skills/plain-language-response/EVAL.md +33 -0
- package/skills/plain-language-response/SKILL.md +194 -0
- package/skills/plain-language-response/contract.json +242 -0
- package/skills/plain-language-response/references/caveat-language.md +25 -0
- package/skills/plain-language-response/references/plain-language-patterns.md +30 -0
- package/skills/plain-language-response/references/stakeholder-friendly-examples.md +30 -0
- package/skills/plain-language-response/references/status-summary-templates.md +50 -0
- package/skills/polycount-budget-checker/EVAL.md +40 -0
- package/skills/polycount-budget-checker/SKILL.md +251 -0
- package/skills/polycount-budget-checker/contract.json +196 -0
- package/skills/polycount-budget-checker/references/instance-counting-rules.md +39 -0
- package/skills/polycount-budget-checker/references/per-mesh-rules.md +43 -0
- package/skills/polycount-budget-checker/references/topology-hints.md +56 -0
- package/skills/pre-handoff-verification/EVAL.md +84 -0
- package/skills/pre-handoff-verification/SKILL.md +252 -0
- package/skills/pre-handoff-verification/contract.json +295 -0
- package/skills/pre-handoff-verification/references/downgrade-rules.md +39 -0
- package/skills/pre-handoff-verification/references/seven-point-checklist.md +46 -0
- package/skills/pre-handoff-verification/references/text-only-mode-handling.md +36 -0
- package/skills/preview-report/EVAL.md +39 -0
- package/skills/preview-report/SKILL.md +254 -0
- package/skills/preview-report/contract.json +141 -0
- package/skills/preview-report/references/handoff-recipients.md +41 -0
- package/skills/preview-report/references/report-schema.md +60 -0
- package/skills/preview-report/references/truth-label-rules.md +36 -0
- package/skills/product-grid-scene-planner/EVAL.md +42 -0
- package/skills/product-grid-scene-planner/SKILL.md +246 -0
- package/skills/product-grid-scene-planner/contract.json +257 -0
- package/skills/product-grid-scene-planner/references/grid-consistency-rules.md +53 -0
- package/skills/product-grid-scene-planner/references/grid-layout-rules.md +42 -0
- package/skills/product-grid-scene-planner/references/grid-output-naming-rules.md +52 -0
- package/skills/product-hero-scene-planner/EVAL.md +33 -0
- package/skills/product-hero-scene-planner/SKILL.md +227 -0
- package/skills/product-hero-scene-planner/contract.json +228 -0
- package/skills/product-hero-scene-planner/references/clarification-patterns.md +24 -0
- package/skills/product-hero-scene-planner/references/cyberpunk-shoe-example.md +28 -0
- package/skills/product-hero-scene-planner/references/product-hero-playbooks.md +22 -0
- package/skills/product-hero-scene-planner/references/scene-plan-schema.md +35 -0
- package/skills/r3f-component-plan/EVAL.md +43 -0
- package/skills/r3f-component-plan/SKILL.md +290 -0
- package/skills/r3f-component-plan/contract.json +171 -0
- package/skills/r3f-component-plan/references/component-shape-patterns.md +54 -0
- package/skills/r3f-component-plan/references/suspense-placement-rules.md +50 -0
- package/skills/r3f-component-plan/references/usegltf-caching-rules.md +53 -0
- package/skills/recipe-fit-assessor/EVAL.md +28 -0
- package/skills/recipe-fit-assessor/SKILL.md +266 -0
- package/skills/recipe-fit-assessor/contract.json +220 -0
- package/skills/recipe-fit-assessor/references/recipe-catalog.md +30 -0
- package/skills/recipe-fit-assessor/references/scope-gap-rules.md +27 -0
- package/skills/recipe-fit-assessor/references/workflow-recommendation-template.md +34 -0
- package/skills/render-export-evidence/EVAL.md +86 -0
- package/skills/render-export-evidence/SKILL.md +339 -0
- package/skills/render-export-evidence/contract.json +206 -0
- package/skills/render-export-evidence/references/attempted-vs-failed-rubric.md +75 -0
- package/skills/render-export-evidence/references/evidence-record-templates.md +143 -0
- package/skills/render-export-evidence/references/truth-label-decision-tree.md +57 -0
- package/skills/resolution-aspect-decision/EVAL.md +26 -0
- package/skills/resolution-aspect-decision/SKILL.md +246 -0
- package/skills/resolution-aspect-decision/contract.json +160 -0
- package/skills/resolution-aspect-decision/references/aspect-defaults.md +41 -0
- package/skills/resolution-aspect-decision/references/print-vs-screen-rules.md +58 -0
- package/skills/resolution-aspect-decision/references/resolution-defaults.md +64 -0
- package/skills/runtime-attempt-report/EVAL.md +26 -0
- package/skills/runtime-attempt-report/SKILL.md +268 -0
- package/skills/runtime-attempt-report/contract.json +176 -0
- package/skills/runtime-attempt-report/references/attempt-record-template.md +49 -0
- package/skills/runtime-attempt-report/references/triage-next-steps.md +15 -0
- package/skills/runtime-attempt-report/references/truth-label-blocks.md +27 -0
- package/skills/runtime-path-picker/EVAL.md +26 -0
- package/skills/runtime-path-picker/SKILL.md +268 -0
- package/skills/runtime-path-picker/contract.json +191 -0
- package/skills/runtime-path-picker/references/anti-patterns.md +14 -0
- package/skills/runtime-path-picker/references/decision-tree.md +29 -0
- package/skills/runtime-path-picker/references/path-constraints-table.md +18 -0
- package/skills/runtime-readiness/EVAL.md +33 -0
- package/skills/runtime-readiness/SKILL.md +218 -0
- package/skills/runtime-readiness/contract.json +190 -0
- package/skills/runtime-readiness/references/blocker-report-template.md +28 -0
- package/skills/runtime-readiness/references/evidence-examples.md +23 -0
- package/skills/runtime-readiness/references/readiness-matrix.md +46 -0
- package/skills/runtime-setup/EVAL.md +33 -0
- package/skills/runtime-setup/SKILL.md +209 -0
- package/skills/runtime-setup/contract.json +171 -0
- package/skills/runtime-setup/references/runtime-option-decision-tree.md +28 -0
- package/skills/runtime-setup/references/setup-summary-template.md +33 -0
- package/skills/runtime-setup/references/source-confidence-labels.md +23 -0
- package/skills/scene-readiness/EVAL.md +33 -0
- package/skills/scene-readiness/SKILL.md +195 -0
- package/skills/scene-readiness/contract.json +198 -0
- package/skills/scene-readiness/references/blocker-catalog.md +24 -0
- package/skills/scene-readiness/references/not-run-handling.md +22 -0
- package/skills/scene-readiness/references/quality-rubric.md +23 -0
- package/skills/scene-readiness/references/verdict-language.md +22 -0
- package/skills/scope-guard/EVAL.md +33 -0
- package/skills/scope-guard/SKILL.md +222 -0
- package/skills/scope-guard/contract.json +171 -0
- package/skills/scope-guard/references/options-language-template.md +51 -0
- package/skills/scope-guard/references/recurring-creep-patterns.md +55 -0
- package/skills/scope-guard/references/scope-addition-classification.md +55 -0
- package/skills/stage-gates/EVAL.md +40 -0
- package/skills/stage-gates/SKILL.md +257 -0
- package/skills/stage-gates/contract.json +490 -0
- package/skills/stage-gates/references/failed-gate-handling.md +42 -0
- package/skills/stage-gates/references/gate-template.md +46 -0
- package/skills/stage-gates/references/stage-decomposition-patterns.md +55 -0
- package/skills/workflow-stage-router/EVAL.md +25 -0
- package/skills/workflow-stage-router/SKILL.md +259 -0
- package/skills/workflow-stage-router/contract.json +150 -0
- package/skills/workflow-stage-router/references/milestone-model.md +20 -0
- package/skills/workflow-stage-router/references/routing-decision-tree.md +32 -0
- package/skills/workflow-stage-router/references/stage-handoff-template.md +37 -0
- package/skills.sh.json +102 -0
|
@@ -0,0 +1,79 @@
|
|
|
1
|
+
# EVAL — blender-brainstorming
|
|
2
|
+
|
|
3
|
+
Status: 1.0.0 text-only baseline
|
|
4
|
+
Scope: Socratic intent exploration; no runtime execution; no artifact claims.
|
|
5
|
+
|
|
6
|
+
## Text-only eval prompt
|
|
7
|
+
|
|
8
|
+
Use blender-brainstorming on a vague non-Blender-user 3D request. Ask 3-5 clarifying questions, propose 2-3 alternative scopings, and produce a confirmed intent summary in plain language. Do not run Blender, do not claim runtime artifacts.
|
|
9
|
+
|
|
10
|
+
## Expected behavior
|
|
11
|
+
|
|
12
|
+
- quotes the original user request verbatim
|
|
13
|
+
- asks 3-5 targeted clarifying questions in one batch
|
|
14
|
+
- proposes 2-3 distinct alternative scopings with trade-offs
|
|
15
|
+
- produces a one-paragraph confirmed intent summary in plain language
|
|
16
|
+
- lists decisions taken vs. deferred
|
|
17
|
+
- runtime status remains `Not Run`; artifact status `Not Produced`
|
|
18
|
+
- hands off to `intent-to-3d-brief-writer` (recommended) or planner
|
|
19
|
+
|
|
20
|
+
## Eval cases
|
|
21
|
+
|
|
22
|
+
| Case | Input | Expected output | Required evidence |
|
|
23
|
+
|---|---|---|---|
|
|
24
|
+
| Vague creative request | "Make me a cool sneaker render" | 5 clarifying questions; 3 alternative scopings (mood, deliverable, audience); user-confirmed intent paragraph | quoted original request, 3-5 questions, 2-3 scopings, paragraph summary |
|
|
25
|
+
| Half-specified request | "I need a cyberpunk shoe hero for web, but I don't know what to do" | 1-3 targeted questions (audience, performance constraints, evidence expectations); 2 scopings; intent paragraph | quoted original, targeted questions, scopings, summary |
|
|
26
|
+
| User refuses clarification | "Just do whatever, I trust you" | Explicit assumption block listing what was assumed; 2 scoping options anyway; low-confidence handoff | assumption list, scoping options, low-confidence label |
|
|
27
|
+
| User keeps expanding scope | "Add this... and also that... and a music video version too" | Stop and ask "lock current scope or keep expanding?"; if locked, summarize | scope-lock prompt, summary of current scope |
|
|
28
|
+
| Already structured brief | User provides subject + mood + deliverable + audience + evidence | Skip brainstorming; route directly to `intent-to-3d-brief-writer` | explicit "skipping brainstorming because brief is complete" message |
|
|
29
|
+
|
|
30
|
+
## Expected evidence/status fields
|
|
31
|
+
|
|
32
|
+
| Field | Required content |
|
|
33
|
+
|---|---|
|
|
34
|
+
| Original user request | Quoted verbatim |
|
|
35
|
+
| Clarifying questions asked | List of 3-5 |
|
|
36
|
+
| Alternative scopings | 2-3 distinct options with trade-offs |
|
|
37
|
+
| Confirmed intent | One paragraph, plain language |
|
|
38
|
+
| Decisions taken | Explicit list |
|
|
39
|
+
| Decisions deferred | Explicit list |
|
|
40
|
+
| Runtime status | `Not Run` |
|
|
41
|
+
| Artifact status | `Not Produced` |
|
|
42
|
+
| Next-skill handoff | Named (typically `intent-to-3d-brief-writer`) |
|
|
43
|
+
|
|
44
|
+
## Pass / Warn / Fail criteria
|
|
45
|
+
|
|
46
|
+
- Pass: all required evidence fields present, alternative scopings distinct (not minor variants), intent summary user-confirmed, no Blender jargon, no runtime claim
|
|
47
|
+
- Warn: clarification depth low (1-2 questions) but explicit assumption notes; OR alternative scopings only 1 with caveats; OR jargon present but brief
|
|
48
|
+
- Fail: zero clarifying questions; OR jumped straight to plan without intent confirmation; OR claimed runtime / artifact production; OR user-facing summary uses dense Blender terms
|
|
49
|
+
|
|
50
|
+
## Common failure modes
|
|
51
|
+
|
|
52
|
+
- drip-feeding questions (one at a time across multiple turns) instead of batching
|
|
53
|
+
- skipping alternative scopings
|
|
54
|
+
- using Blender jargon ("topology", "subsurf modifier", "BSDF shader") in the intent summary aimed at non-Blender users
|
|
55
|
+
- claiming "I'll now set up Blender" or "I'll install the Connector" — these are out of scope for this skill
|
|
56
|
+
|
|
57
|
+
## Evidence expectations
|
|
58
|
+
|
|
59
|
+
- conversation log includes verbatim user request
|
|
60
|
+
- batch of 3-5 questions visible in one turn
|
|
61
|
+
- alternative scopings clearly enumerated
|
|
62
|
+
- confirmed intent paragraph echoed back to user before handoff
|
|
63
|
+
- explicit `Not Run` / `Not Produced` status
|
|
64
|
+
|
|
65
|
+
## Sample passing response outline
|
|
66
|
+
|
|
67
|
+
- "What I heard you ask for: <quote>"
|
|
68
|
+
- "What I need to know: <3-5 questions>"
|
|
69
|
+
- "Three ways we could scope this: <option A / B / C with trade-offs>"
|
|
70
|
+
- "Confirmed intent (your scope: <X>): <one paragraph>"
|
|
71
|
+
- "Decisions taken: <list>. Deferred: <list>."
|
|
72
|
+
- "Next step: hand off to `intent-to-3d-brief-writer`. Runtime status: Not Run."
|
|
73
|
+
|
|
74
|
+
## Sample failing response outline
|
|
75
|
+
|
|
76
|
+
- Skipped clarification, dove straight into a multi-paragraph scene plan
|
|
77
|
+
- No alternative scopings offered
|
|
78
|
+
- Used "topology", "BSDF", "subsurf" without explanation
|
|
79
|
+
- Claimed "I've set up the Blender connector" when no setup happened
|
|
@@ -0,0 +1,228 @@
|
|
|
1
|
+
---
|
|
2
|
+
name: blender-brainstorming
|
|
3
|
+
description: Socratic exploration of a non-Blender-user's 3D intent before scene planning. Use when the request is vague, has multiple plausible interpretations, or needs creative scoping.
|
|
4
|
+
---
|
|
5
|
+
|
|
6
|
+
# blender-brainstorming
|
|
7
|
+
|
|
8
|
+
## Purpose
|
|
9
|
+
|
|
10
|
+
Run a structured Socratic brainstorming pass on a vague non-Blender-user 3D request before any scene planner, runtime readiness, or evidence skill is invoked. Surface the user's actual intent, missing constraints, and decision points so downstream skills do not waste effort on the wrong direction.
|
|
11
|
+
|
|
12
|
+
Inspired by the `/brainstorming` skill in Anthropic Superpowers, adapted for the Blender / non-Blender-user domain.
|
|
13
|
+
|
|
14
|
+
## Quick start
|
|
15
|
+
|
|
16
|
+
- ask 3-5 minimum clarifying questions
|
|
17
|
+
- propose 2-3 alternative scopings
|
|
18
|
+
- end with a one-paragraph intent summary the user can confirm
|
|
19
|
+
- hand off to `intent-to-3d-brief-writer` once intent is confirmed
|
|
20
|
+
|
|
21
|
+
## When to use
|
|
22
|
+
|
|
23
|
+
- the user request is short, ambiguous, or missing critical context (subject, mood, deliverable, evidence expectations)
|
|
24
|
+
- multiple plausible interpretations exist (e.g. "make me a cool product render" — what product, what mood, web or print, GLB or render?)
|
|
25
|
+
- the user is non-Blender-technical and may not realize what BlendOps must know to plan well
|
|
26
|
+
- before any planner skill (`product-hero-scene-planner`, `camera-composition`, …)
|
|
27
|
+
|
|
28
|
+
## When not to use
|
|
29
|
+
<!-- blendops:generated:start exclusions -->
|
|
30
|
+
- the user already provided a structured brief (route directly to intent-to-3d-brief-writer to lock the brief before any planner runs) → use `../intent-to-3d-brief-writer/SKILL.md` (route directly to intent-to-3d-brief-writer to lock the brief before any planner runs)
|
|
31
|
+
- the user explicitly asked to skip clarification (record explicit assumptions and route to intent-to-3d-brief-writer) → use `../intent-to-3d-brief-writer/SKILL.md` (record explicit assumptions and route to intent-to-3d-brief-writer)
|
|
32
|
+
- the user pushes for immediate execution rather than exploration → use `../runtime-readiness/SKILL.md` (route to runtime-readiness)
|
|
33
|
+
<!-- blendops:generated:end exclusions -->
|
|
34
|
+
|
|
35
|
+
## Trigger phrases
|
|
36
|
+
<!-- blendops:generated:start triggers -->
|
|
37
|
+
- "I want a 3D render of..." (vague)"
|
|
38
|
+
- "Make me something cool with..."
|
|
39
|
+
- "Help me plan a Blender scene" (no constraints given)"
|
|
40
|
+
- "I'm not sure what I want exactly..."
|
|
41
|
+
- "Brainstorm with me about..."
|
|
42
|
+
<!-- blendops:generated:end triggers -->
|
|
43
|
+
|
|
44
|
+
## Prerequisites / readiness
|
|
45
|
+
|
|
46
|
+
- ability to ask the user questions in turn
|
|
47
|
+
- no Blender runtime required
|
|
48
|
+
- works in text-only mode
|
|
49
|
+
|
|
50
|
+
## Input schema
|
|
51
|
+
|
|
52
|
+
### Required inputs
|
|
53
|
+
|
|
54
|
+
- the original user request (verbatim)
|
|
55
|
+
- target deliverable hint if any (image / GLB / web hero / iteration loop)
|
|
56
|
+
|
|
57
|
+
### Optional inputs
|
|
58
|
+
|
|
59
|
+
- existing references (mood boards, brand color hints)
|
|
60
|
+
- known performance / time / scope constraints
|
|
61
|
+
- target audience for the final output
|
|
62
|
+
|
|
63
|
+
### Assumptions to confirm
|
|
64
|
+
|
|
65
|
+
- the user has time for a single 3-5 batched question clarification round
|
|
66
|
+
- the user wants a confirmed intent summary before brief locking and planning
|
|
67
|
+
- truth states are locked at `Runtime: Not Run` and `Artifact: Not Produced` (conversation-only; downgrade-only, never upgrade)
|
|
68
|
+
|
|
69
|
+
## Output schema
|
|
70
|
+
|
|
71
|
+
### Primary output
|
|
72
|
+
|
|
73
|
+
- user-confirmed one-paragraph intent summary in plain language (no Blender jargon)
|
|
74
|
+
- explicit list of decisions taken vs. deferred
|
|
75
|
+
- named handoff target (`intent-to-3d-brief-writer` to lock the brief, `runtime-readiness` if execution is pushed, or `plain-language-response` if blocked)
|
|
76
|
+
|
|
77
|
+
### Secondary output
|
|
78
|
+
|
|
79
|
+
- batched list of 3-5 clarifying questions asked + user answers
|
|
80
|
+
- 2-3 alternative scopings considered (and why one was chosen)
|
|
81
|
+
|
|
82
|
+
### Evidence / caveat output
|
|
83
|
+
|
|
84
|
+
- `Runtime: Not Run` | `Artifact: Not Produced` (downgrade-only, never upgrade)
|
|
85
|
+
- any unresolved ambiguity flagged as an explicit blocker for `intent-to-3d-brief-writer`
|
|
86
|
+
- explicit statement that this session is conversation-only and no Blender installation or execution happened
|
|
87
|
+
|
|
88
|
+
## Required laws
|
|
89
|
+
|
|
90
|
+
- ../../laws/non-blender-user-language.md
|
|
91
|
+
- ../../laws/evidence-before-done.md
|
|
92
|
+
- ../../laws/no-arbitrary-python-interface.md
|
|
93
|
+
- ../../laws/official-runtime-only.md
|
|
94
|
+
|
|
95
|
+
## Official runtime boundary
|
|
96
|
+
|
|
97
|
+
This skill does not install Blender, run Blender, or configure any MCP path. It is text-only conversation.
|
|
98
|
+
|
|
99
|
+
If the user pushes for execution during brainstorming, route to `runtime-readiness` and do not attempt runtime work from this skill.
|
|
100
|
+
|
|
101
|
+
## Operating procedure
|
|
102
|
+
|
|
103
|
+
1. Read the user's original request verbatim. Do not paraphrase yet.
|
|
104
|
+
2. Identify the core facts needed downstream (subject, mood, deliverable type, audience, evidence expectations, and open creative constraints).
|
|
105
|
+
3. Formulate 3-5 short clarifying questions in plain language and ask all of them in a single batch (never drip-feed one question at a time).
|
|
106
|
+
4. Propose 2-3 distinct alternative scopings (e.g., "minimal hero image" vs. "interactive web-ready 3D model" vs. "iteration-friendly concept pass").
|
|
107
|
+
5. Ask the user to pick one scoping or combine elements, and confirm the resulting one-paragraph intent summary in plain language (no Blender jargon).
|
|
108
|
+
6. List decisions taken vs. deferred and record any remaining ambiguity as an explicit blocker.
|
|
109
|
+
7. Maintain `Runtime: Not Run` and `Artifact: Not Produced` (downgrade-only, never upgrade).
|
|
110
|
+
8. Hand off to `intent-to-3d-brief-writer` to lock the structured brief downstream before any planner skill is invoked (or route to `runtime-readiness` if the user pushes for execution mid-brainstorm).
|
|
111
|
+
|
|
112
|
+
## Decision tree
|
|
113
|
+
|
|
114
|
+
- request is vague or missing key facts → ask 3-5 batched clarifying questions covering missing facts and creative trade-offs, then propose 2-3 alternative scopings.
|
|
115
|
+
- request is already a complete structured brief → skip brainstorming and route directly to `intent-to-3d-brief-writer`.
|
|
116
|
+
- user pushes for execution mid-brainstorm → pause brainstorming and route immediately to `runtime-readiness`.
|
|
117
|
+
- user refuses to answer clarifying questions → record explicit assumptions and unresolved blockers, keep `Runtime: Not Run` and `Artifact: Not Produced` (never invent confidence labels), and hand off to `intent-to-3d-brief-writer`.
|
|
118
|
+
|
|
119
|
+
## Playbooks
|
|
120
|
+
|
|
121
|
+
- Playbook A: "Vague creative request" — ask 5 batched clarifying questions, propose 3 alternative scopings, confirm one-paragraph plain-language summary, hand off to `intent-to-3d-brief-writer`.
|
|
122
|
+
- Playbook B: "Partially specified request" — ask 3-4 batched clarifying questions on remaining gaps and scoping trade-offs, propose 2-3 alternative scopings, confirm one-paragraph summary, hand off to `intent-to-3d-brief-writer`.
|
|
123
|
+
- Playbook C: "User refuses clarification" — propose 2-3 alternative scopings with explicit assumption notes and blockers, keep `Runtime: Not Run` and `Artifact: Not Produced` without invented confidence labels, and hand off to `intent-to-3d-brief-writer`.
|
|
124
|
+
|
|
125
|
+
## Mode handling
|
|
126
|
+
|
|
127
|
+
### Text-only mode
|
|
128
|
+
|
|
129
|
+
- conversation only; never runs Blender or installs anything
|
|
130
|
+
- `Runtime: Not Run`
|
|
131
|
+
- `Artifact: Not Produced` (downgrade-only, never upgrade)
|
|
132
|
+
|
|
133
|
+
### Runtime-ready mode
|
|
134
|
+
|
|
135
|
+
- brainstorming remains strictly conversation-only (`Runtime: Not Run`, `Artifact: Not Produced`)
|
|
136
|
+
- if the user pushes for execution instead of scoping, route to `runtime-readiness`
|
|
137
|
+
|
|
138
|
+
### Blocked runtime mode
|
|
139
|
+
|
|
140
|
+
- brainstorming still runs as text-only conversation (`Runtime: Not Run`, `Artifact: Not Produced`)
|
|
141
|
+
- note that downstream runtime execution is blocked and route to `plain-language-response` to explain blocked status in plain language
|
|
142
|
+
|
|
143
|
+
## Validation checklist
|
|
144
|
+
|
|
145
|
+
- [ ] original user request quoted verbatim
|
|
146
|
+
- [ ] 3-5 clarifying questions asked in a single batch (no drip-feeding)
|
|
147
|
+
- [ ] 2-3 alternative scopings proposed
|
|
148
|
+
- [ ] user-confirmed one-paragraph intent summary produced in plain language (no Blender jargon)
|
|
149
|
+
- [ ] decisions taken vs. deferred listed explicitly
|
|
150
|
+
- [ ] truth states explicitly set to `Runtime: Not Run` and `Artifact: Not Produced` (downgrade-only, never upgrade)
|
|
151
|
+
- [ ] no invented confidence or status labels used
|
|
152
|
+
- [ ] named handoff target matches role split (`intent-to-3d-brief-writer`, `runtime-readiness`, or `plain-language-response`)
|
|
153
|
+
- [ ] no statement or question implies BlendOps installs or executes Blender
|
|
154
|
+
- [ ] unresolved ambiguity flagged as an explicit blocker
|
|
155
|
+
|
|
156
|
+
## Pass / Warn / Fail rubric
|
|
157
|
+
|
|
158
|
+
| Category | Pass | Warn | Fail |
|
|
159
|
+
|---|---|---|---|
|
|
160
|
+
| Clarification & scoping | 3-5 batched questions + 2-3 distinct alternative scopings | User declined questions; 2-3 scopings and explicit assumptions recorded | Drip-feed questioning (<3 batched questions) or <2 alternative scopings |
|
|
161
|
+
| Intent summary | User-confirmed one-paragraph summary in plain language (no Blender jargon) | Plain-language one-paragraph summary awaiting final user confirmation | Jargon-heavy, multi-paragraph, or missing summary |
|
|
162
|
+
| Runtime & truth states | `Runtime: Not Run`, `Artifact: Not Produced` (downgrade-only); no install/execution claims; no invented labels | Truth states present but blocker note could be clearer | Upgrades truth states, claims install/execution, or invents confidence/status labels |
|
|
163
|
+
| Role boundary & handoff | Named handoff to `intent-to-3d-brief-writer`, `runtime-readiness`, or `plain-language-response` | Named valid handoff with minor caveat omission | No handoff, or bypasses `intent-to-3d-brief-writer` to jump straight into scene planning |
|
|
164
|
+
|
|
165
|
+
## Failure handling
|
|
166
|
+
|
|
167
|
+
- user gives one-word answers: record explicit assumption notes for remaining gaps, present the 2-3 alternative scopings, and ask the user to confirm the one-paragraph summary.
|
|
168
|
+
- user contradicts themselves: surface the contradiction in plain language and ask which direction to lock in the summary.
|
|
169
|
+
- user wants execution mid-brainstorm: pause brainstorming, route to `runtime-readiness`, and keep `Runtime: Not Run` and `Artifact: Not Produced`.
|
|
170
|
+
|
|
171
|
+
## Troubleshooting
|
|
172
|
+
|
|
173
|
+
- if the user keeps expanding scope: summarize the current scope in plain language, present 2-3 bounded scoping options, and ask whether to lock one for `intent-to-3d-brief-writer`.
|
|
174
|
+
- if key facts remain unanswered after the batched questions: record the missing facts as explicit blockers and assumptions while keeping `Runtime: Not Run` and `Artifact: Not Produced` (do not invent custom status or confidence labels).
|
|
175
|
+
|
|
176
|
+
## Best practices
|
|
177
|
+
|
|
178
|
+
- ask questions in batches, not one at a time
|
|
179
|
+
- propose alternatives the user might not have considered
|
|
180
|
+
- confirm intent before handoff (user reads summary, says "yes/no")
|
|
181
|
+
|
|
182
|
+
## Good examples
|
|
183
|
+
|
|
184
|
+
- "User asked: 'Make me a cool sneaker render.' Brainstorming surfaces: cyberpunk vs. minimalist mood; web GLB vs. marketing render; iteration vs. one-shot. User picks cyberpunk + web GLB. Intent locked."
|
|
185
|
+
|
|
186
|
+
## Bad examples
|
|
187
|
+
|
|
188
|
+
- "User asked vague request. Skipping brainstorming and jumping straight to scene planner with assumed parameters." (overclaim risk; user will be surprised)
|
|
189
|
+
|
|
190
|
+
## User-facing response template
|
|
191
|
+
|
|
192
|
+
- **What I heard you ask for:** (verbatim quote)
|
|
193
|
+
- **Questions to narrow this down:** (3-5 batched questions in plain language)
|
|
194
|
+
- **Alternative ways we can scope this:** (2-3 distinct options)
|
|
195
|
+
- **Confirmed intent summary:** (one plain-language paragraph, no Blender jargon, confirmed by user)
|
|
196
|
+
- **Decisions taken vs. deferred & blockers:** (explicit list)
|
|
197
|
+
- **Status:** `Runtime: Not Run` | `Artifact: Not Produced` (conversation-only)
|
|
198
|
+
- **Next step:** Hand off to `intent-to-3d-brief-writer` (or `runtime-readiness` / `plain-language-response` when applicable)
|
|
199
|
+
|
|
200
|
+
## Anti-patterns
|
|
201
|
+
|
|
202
|
+
- asking 1 question at a time (drip-feed questioning) instead of 3-5 batched questions
|
|
203
|
+
- skipping the 2-3 alternative scopings
|
|
204
|
+
- using Blender jargon in the one-paragraph intent summary
|
|
205
|
+
- claiming the brainstorming session installed or executed Blender, or upgrading `Runtime: Not Run` / `Artifact: Not Produced`
|
|
206
|
+
- inventing custom confidence or status labels (such as `verified-read: low` or `Unknown`)
|
|
207
|
+
- blurring role boundaries by planning the scene directly or bypassing `intent-to-3d-brief-writer`
|
|
208
|
+
|
|
209
|
+
## Cross-skill handoff
|
|
210
|
+
<!-- blendops:generated:start handoffs -->
|
|
211
|
+
- [progression] Lock the confirmed brief downstream → `../intent-to-3d-brief-writer/SKILL.md`
|
|
212
|
+
- [progression] User pushes for execution mid-brainstorm → `../runtime-readiness/SKILL.md`
|
|
213
|
+
- [progression] Explain blocked runtime status to user → `../plain-language-response/SKILL.md`
|
|
214
|
+
<!-- blendops:generated:end handoffs -->
|
|
215
|
+
|
|
216
|
+
## Non-goals
|
|
217
|
+
|
|
218
|
+
- runtime setup
|
|
219
|
+
- artifact production
|
|
220
|
+
- deciding the final scene plan (that's the planner's job)
|
|
221
|
+
- one-shot interpretation of vague requests
|
|
222
|
+
|
|
223
|
+
## References
|
|
224
|
+
|
|
225
|
+
- Inspired by Anthropic Superpowers `/brainstorming` skill (https://github.com/anthropics/skills)
|
|
226
|
+
- BlendOps law: ../../laws/non-blender-user-language.md
|
|
227
|
+
- BlendOps law: ../../laws/evidence-before-done.md
|
|
228
|
+
- Skill system: ../../docs/skill-system.md
|
|
@@ -0,0 +1,115 @@
|
|
|
1
|
+
{
|
|
2
|
+
"$schema": "../../schemas/skill-contract.schema.json",
|
|
3
|
+
"id": "blender-brainstorming",
|
|
4
|
+
"version": "1.0.0",
|
|
5
|
+
"category": "process-discipline",
|
|
6
|
+
"role": "pre-brief-ideation-and-clarification",
|
|
7
|
+
"purpose": "Socratic exploration of a non-Blender-user's 3D intent before scene planning. Use when the request is vague, has multiple plausible interpretations, or needs creative scoping.",
|
|
8
|
+
"capabilityTier": "L0",
|
|
9
|
+
"triggers": [
|
|
10
|
+
"I want a 3D render of...\" (vague)",
|
|
11
|
+
"Make me something cool with...",
|
|
12
|
+
"Help me plan a Blender scene\" (no constraints given)",
|
|
13
|
+
"I'm not sure what I want exactly...",
|
|
14
|
+
"Brainstorm with me about..."
|
|
15
|
+
],
|
|
16
|
+
"exclusions": [
|
|
17
|
+
{
|
|
18
|
+
"condition": "the user already provided a structured brief (route directly to intent-to-3d-brief-writer to lock the brief before any planner runs)",
|
|
19
|
+
"delegateTo": "intent-to-3d-brief-writer",
|
|
20
|
+
"reason": "route directly to intent-to-3d-brief-writer to lock the brief before any planner runs"
|
|
21
|
+
},
|
|
22
|
+
{
|
|
23
|
+
"condition": "the user explicitly asked to skip clarification (record explicit assumptions and route to intent-to-3d-brief-writer)",
|
|
24
|
+
"delegateTo": "intent-to-3d-brief-writer",
|
|
25
|
+
"reason": "record explicit assumptions and route to intent-to-3d-brief-writer"
|
|
26
|
+
},
|
|
27
|
+
{
|
|
28
|
+
"condition": "the user pushes for immediate execution rather than exploration",
|
|
29
|
+
"delegateTo": "runtime-readiness",
|
|
30
|
+
"reason": "route to runtime-readiness"
|
|
31
|
+
}
|
|
32
|
+
],
|
|
33
|
+
"inputs": {
|
|
34
|
+
"required": [
|
|
35
|
+
{
|
|
36
|
+
"name": "initial_user_idea",
|
|
37
|
+
"type": "string",
|
|
38
|
+
"description": "Vague, exploratory, or open-ended 3D concept from the user"
|
|
39
|
+
}
|
|
40
|
+
],
|
|
41
|
+
"optional": [
|
|
42
|
+
{
|
|
43
|
+
"name": "brand_or_mood_references",
|
|
44
|
+
"type": "array",
|
|
45
|
+
"description": "Optional mood keywords, reference products, or target audience notes"
|
|
46
|
+
}
|
|
47
|
+
]
|
|
48
|
+
},
|
|
49
|
+
"outputs": {
|
|
50
|
+
"primary": [
|
|
51
|
+
"divergent_concept_options",
|
|
52
|
+
"clarifying_questions_shortlist",
|
|
53
|
+
"recommended_direction_for_brief"
|
|
54
|
+
],
|
|
55
|
+
"truthStatePolicy": {
|
|
56
|
+
"runtimeDefault": "Not Run",
|
|
57
|
+
"artifactDefault": "Not Produced",
|
|
58
|
+
"mutationRule": "downgrade_only"
|
|
59
|
+
}
|
|
60
|
+
},
|
|
61
|
+
"handoffs": [
|
|
62
|
+
{
|
|
63
|
+
"target": "intent-to-3d-brief-writer",
|
|
64
|
+
"type": "progression",
|
|
65
|
+
"label": "Lock the confirmed brief downstream",
|
|
66
|
+
"when": {
|
|
67
|
+
"all": [
|
|
68
|
+
{
|
|
69
|
+
"field": "goal.clarity",
|
|
70
|
+
"operator": "equals",
|
|
71
|
+
"value": "clear"
|
|
72
|
+
},
|
|
73
|
+
{
|
|
74
|
+
"field": "brief.locked",
|
|
75
|
+
"operator": "equals",
|
|
76
|
+
"value": false
|
|
77
|
+
}
|
|
78
|
+
]
|
|
79
|
+
}
|
|
80
|
+
},
|
|
81
|
+
{
|
|
82
|
+
"target": "runtime-readiness",
|
|
83
|
+
"type": "progression",
|
|
84
|
+
"label": "User pushes for execution mid-brainstorm",
|
|
85
|
+
"when": {
|
|
86
|
+
"all": [
|
|
87
|
+
{
|
|
88
|
+
"field": "runtime.path",
|
|
89
|
+
"operator": "not_equals",
|
|
90
|
+
"value": "unselected"
|
|
91
|
+
},
|
|
92
|
+
{
|
|
93
|
+
"field": "runtime.verifiedReady",
|
|
94
|
+
"operator": "equals",
|
|
95
|
+
"value": false
|
|
96
|
+
}
|
|
97
|
+
]
|
|
98
|
+
}
|
|
99
|
+
},
|
|
100
|
+
{
|
|
101
|
+
"target": "plain-language-response",
|
|
102
|
+
"type": "progression",
|
|
103
|
+
"label": "Explain blocked runtime status to user",
|
|
104
|
+
"when": {
|
|
105
|
+
"all": [
|
|
106
|
+
{
|
|
107
|
+
"field": "stage.current",
|
|
108
|
+
"operator": "equals",
|
|
109
|
+
"value": "final_communication"
|
|
110
|
+
}
|
|
111
|
+
]
|
|
112
|
+
}
|
|
113
|
+
}
|
|
114
|
+
]
|
|
115
|
+
}
|
|
@@ -0,0 +1,43 @@
|
|
|
1
|
+
# Alternative Scoping Templates
|
|
2
|
+
|
|
3
|
+
After clarifying questions are answered, propose 2-3 distinct scopings. Distinct means **the trade-offs are visible**, not minor variants.
|
|
4
|
+
|
|
5
|
+
## Standard 3-tier scoping
|
|
6
|
+
|
|
7
|
+
For most product-hero / scene-render requests:
|
|
8
|
+
|
|
9
|
+
### Scoping A — Minimal hero render
|
|
10
|
+
|
|
11
|
+
- Single render image, one camera angle.
|
|
12
|
+
- One mood lock (no iteration on mood).
|
|
13
|
+
- Acceptance: subject is recognizable + mood reads.
|
|
14
|
+
- Time-to-handoff: short.
|
|
15
|
+
- Trade-off: no GLB, no web handoff, no iteration.
|
|
16
|
+
|
|
17
|
+
### Scoping B — Full web-ready GLB + render
|
|
18
|
+
|
|
19
|
+
- Render image + GLB export.
|
|
20
|
+
- Web performance budget enforced.
|
|
21
|
+
- Acceptance: render passes mood check + GLB passes size + import check.
|
|
22
|
+
- Time-to-handoff: medium.
|
|
23
|
+
- Trade-off: poly budget tightens lookdev choices.
|
|
24
|
+
|
|
25
|
+
### Scoping C — Iteration-friendly text-only plan first
|
|
26
|
+
|
|
27
|
+
- Scene plan + composition + lighting + material plans.
|
|
28
|
+
- No render, no GLB.
|
|
29
|
+
- Acceptance: plans are reviewable on paper.
|
|
30
|
+
- Time-to-handoff: shortest.
|
|
31
|
+
- Trade-off: no visual proof, runtime status stays `Not Run`.
|
|
32
|
+
|
|
33
|
+
## When to deviate from the 3-tier
|
|
34
|
+
|
|
35
|
+
- User has a brand asset library → Scoping A may extend to "import branded model + minimal render".
|
|
36
|
+
- User has Hyper3D / Hunyuan3D access (Path 2) → add Scoping D "generative subject + procedural environment".
|
|
37
|
+
- User wants AR / USDZ → add Scoping E "GLB + AR handoff" (route through `glb-web-handoff` for caveats).
|
|
38
|
+
|
|
39
|
+
## Anti-patterns
|
|
40
|
+
|
|
41
|
+
- 3 scopings that only differ in mood. Mood was already in question 2.
|
|
42
|
+
- Scopings that all need runtime — iteration scopings should always include a text-only option.
|
|
43
|
+
- Scopings that ignore the audience answer (question 4). Audience drives evidence expectations.
|
|
@@ -0,0 +1,29 @@
|
|
|
1
|
+
# Clarification Patterns
|
|
2
|
+
|
|
3
|
+
Five-fact framework for batched clarifying questions. Ask all missing facts in **one batch**, never drip-feed.
|
|
4
|
+
|
|
5
|
+
## The 5 minimum facts a downstream planner needs
|
|
6
|
+
|
|
7
|
+
| # | Fact | Example question |
|
|
8
|
+
|---|---|---|
|
|
9
|
+
| 1 | **Subject** | "What's the hero subject — a product, a character, an environment, an abstract scene?" |
|
|
10
|
+
| 2 | **Mood / style** | "What mood do you want — minimalist studio, cyberpunk neon, sunset warmth, brutalist concrete, something else?" |
|
|
11
|
+
| 3 | **Deliverable type** | "Do you need a render image, a GLB for web, both, or just a planning iteration?" |
|
|
12
|
+
| 4 | **Audience** | "Who's the final audience — a marketing team, a non-technical client, a tech artist, mixed?" |
|
|
13
|
+
| 5 | **Evidence expectations** | "Are you OK with a text-only plan first (`Not Run`), or do you need actual rendered output (`Verified`) by a deadline?" |
|
|
14
|
+
|
|
15
|
+
## Question batching rules
|
|
16
|
+
|
|
17
|
+
- **One batch per turn.** All missing-fact questions in a single message.
|
|
18
|
+
- **Plain language.** No jargon (no "topology", "BSDF", "subsurf").
|
|
19
|
+
- **Concrete examples in each question.** Helps the user choose without explaining vocabulary.
|
|
20
|
+
- **Optional facts** (brand color, performance budget, references) go in a SECOND batch only if the first batch reveals they matter.
|
|
21
|
+
|
|
22
|
+
## When to skip clarification
|
|
23
|
+
|
|
24
|
+
- 4-5 of 5 facts are already in the user's request → skip brainstorming, route directly to `intent-to-3d-brief-writer`.
|
|
25
|
+
- User explicitly says "just go" → record explicit assumptions, low-confidence handoff.
|
|
26
|
+
|
|
27
|
+
## When to push back
|
|
28
|
+
|
|
29
|
+
- User answers everything with "I don't know, you decide" → propose 2-3 default scopings (see `alternative-scoping-templates.md`) and ask the user to pick one. **Do not invent answers silently.**
|
|
@@ -0,0 +1,25 @@
|
|
|
1
|
+
# Intent Summary Style Guide
|
|
2
|
+
|
|
3
|
+
After the user picks a scoping, write a **one-paragraph confirmed intent summary** that the user echoes back to confirm.
|
|
4
|
+
|
|
5
|
+
## Required style
|
|
6
|
+
|
|
7
|
+
- **Plain language.** No Blender jargon.
|
|
8
|
+
- **One paragraph, ≤5 sentences.**
|
|
9
|
+
- **Subject + mood + deliverable + audience + evidence expectations** all named.
|
|
10
|
+
- **No vague modifiers** ("cool", "nice", "awesome"). Use the user's own descriptive words instead.
|
|
11
|
+
- **End with the next-skill handoff statement.**
|
|
12
|
+
|
|
13
|
+
## Template
|
|
14
|
+
|
|
15
|
+
> You want a [SUBJECT] in [MOOD/STYLE], delivered as [DELIVERABLES], for [AUDIENCE]. Acceptance is [OBSERVABLE CRITERIA]. Evidence expectation is [`Not Run` / `Produced` / `Verified`]. Decisions deferred to planners: [LIST]. Next step: hand off to [NEXT SKILL] for [PURPOSE].
|
|
16
|
+
|
|
17
|
+
## Filled example
|
|
18
|
+
|
|
19
|
+
> You want a floating cyberpunk sneaker in a glossy-dark-floor neon-noir mood, delivered as a render image **and** a GLB for web embed, for your marketing team (non-Blender technical). Acceptance is: subject framing is centered with negative space for headline, neon mood is visible, GLB ≤ 10MB. Evidence expectation: `Verified` for the render, `Produced` for GLB pending web import test. Decisions deferred to planners: exact lighting placement (lighting planner), exact camera angle (composition planner), shoe surface material (material planner). Next step: hand off to `intent-to-3d-brief-writer` to lock the 8-slot brief.
|
|
20
|
+
|
|
21
|
+
## Anti-patterns
|
|
22
|
+
|
|
23
|
+
- "Make a cool cyberpunk render of a shoe." (vague modifier "cool", no audience, no evidence expectation, no handoff)
|
|
24
|
+
- "User wants subsurf-shaded BSDF cyberpunk subject with topology budget." (Blender jargon for non-Blender audience)
|
|
25
|
+
- "We'll figure out the rest as we go." (no decisions deferred list, no observable acceptance criteria)
|
|
@@ -0,0 +1,35 @@
|
|
|
1
|
+
# EVAL — blender-recipe-decomposer
|
|
2
|
+
|
|
3
|
+
## Text-only eval prompt
|
|
4
|
+
User intent: "I need a hero render of the shoe + 4 color variants + a 3-second orbit animation for the web". Apply blender-recipe-decomposer.
|
|
5
|
+
|
|
6
|
+
## Expected behavior
|
|
7
|
+
- 3 recipes named (hero render / variants grid / orbit animation)
|
|
8
|
+
- Brief seed + stop condition per recipe
|
|
9
|
+
- Order + dependencies explicit (variants and animation depend on hero scene)
|
|
10
|
+
- Shared assets identified
|
|
11
|
+
- Start recipe recommended
|
|
12
|
+
- Each recipe hands off to intent-to-3d-brief-writer
|
|
13
|
+
|
|
14
|
+
## Pass / Warn / Fail criteria
|
|
15
|
+
- Pass: 3 cohesive recipes + names + briefs + stops + order + dependencies + shared assets
|
|
16
|
+
- Warn: stop conditions soft or dependencies unclear
|
|
17
|
+
- Fail: one mega-recipe; over-split (5+ fragments); missing stops
|
|
18
|
+
|
|
19
|
+
## Common failure modes
|
|
20
|
+
- single mega-recipe
|
|
21
|
+
- recipe overlap in scope
|
|
22
|
+
- skipping dependency analysis
|
|
23
|
+
- naming recipes by process not deliverable
|
|
24
|
+
|
|
25
|
+
## Sample passing response outline
|
|
26
|
+
- User intent + deliverable enumeration
|
|
27
|
+
- 3-row recipe table
|
|
28
|
+
- Shared assets list
|
|
29
|
+
- Recommended start
|
|
30
|
+
- Handoff to intent-to-3d-brief-writer
|
|
31
|
+
|
|
32
|
+
## Sample failing response outline
|
|
33
|
+
- "Do all in one go"
|
|
34
|
+
- No stop conditions
|
|
35
|
+
- No dependencies
|