unity-perception
Skill7 starsChanged 7 months agoMITView on GitHub
Use when users want to get a summary, overview, dependency report, or export of the current scene state.
In this repository34 other files it gives its agents.
- unity-mcp-skillsSkill
- unity-animatorSkill
- unity-assetSkill
- unity-cameraSkill
- unity-cinemachineSkill
- unity-cleanerSkill
- unity-componentSkill
- unity-consoleSkill
- unity-editorSkill
- unity-eventSkill
- unity-gameobjectSkill
- unity-importerSkill
- unity-lightSkill
- unity-materialSkill
- unity-navmeshSkill
- unity-optimizationSkill
- unity-packageSkill
- unity-physicsSkill
- unity-prefabSkill
- unity-probuilderSkill
- unity-projectSkill
- unity-sceneSkill
- unity-scriptableobjectSkill
- unity-scriptSkill
- unity-shaderSkill
- unity-smartSkill
- unity-terrainSkill
- unity-testSkill
- unity-timelineSkill
- unity-toolingSkill
- unity-uiSkill
- unity-uitoolkitSkill
- unity-validationSkill
- unity-xrSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 10
Use this module for read-only scene and project analysis.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Unity Perception Skills
Overview
Use this module for read-only scene and project analysis. It is available in Semi-Auto by default.
Common Mistakes
DO NOT (common hallucinations):
perception_analyze,perception_scan, andperception_describedo not existscene_contextis noteditor_get_context: it exports hierarchy/components/references, while editor context focuses on current editor statescene_analyze,scene_health_check,scene_contract_validate,scene_component_stats,scene_find_hotspots, andproject_stack_detectbelong to this module even if the prefix looks likescene_*orproject_*
Routing:
- Current selection/play-mode/editor state ->
editor_get_context - Object search by name/path ->
scene_find_objectsorgameobject_find - Script dependency closure ->
script_dependency_graph
Skills
Scene Health and Summary
| Skill | Use | Key parameters |
scene_analyze | Combined scene + project analysis | topComponentsLimit?, issueLimit?, deepHierarchyThreshold?, largeChildCountThreshold? |
scene_health_check | Read-only health report | issueLimit?, deepHierarchyThreshold?, largeChildCountThreshold? |
scene_summarize | Structured scene summary | includeComponentStats?, topComponentsLimit? |
scene_component_stats | Component and facility stats | topComponentsLimit? |
scene_find_hotspots | Deep hierarchy / large group / empty node hotspots | thresholds + maxResults? |
scene_tag_layer_stats | Tag/layer usage | none |
scene_performance_hints | Prioritized optimization hints | none |
Scene Snapshots and Exports
| Skill | Use | Key parameters |
scene_diff | Capture or compare lightweight snapshots | snapshotJson? |
hierarchy_describe | Return text hierarchy tree | maxDepth?, includeInactive?, maxItemsPerLevel? |
scene_context | Export hierarchy, components, references | maxDepth?, maxObjects?, rootPath?, includeValues?, includeReferences?, includeCodeDeps? |
scene_export_report | Save markdown scene report | savePath?, maxDepth?, maxObjects? |
scene_dependency_analyze | Analyze impact / dependency graph in-scene | targetPath?, savePath? |
Project and Script Analysis
| Skill | Use | Key parameters |
project_stack_detect | Detect pipeline, input, UI, packages, tests, folders | none |
script_analyze | Analyze one MonoBehaviour / ScriptableObject / user class by class name | scriptName, includePrivate? |
script_dependency_graph | N-hop dependency closure for one script class name | scriptName, maxHops?, includeDetails? |
Spatial and Material Queries
| Skill | Use | Key parameters |
scene_spatial_query | Find objects near a point or object | x/y/z?, radius?, nearObject?, componentFilter?, maxResults? |
scene_materials | Summarize scene materials and shaders | includeProperties? |
scene_contract_validate | Validate default roots/tags/layers/UI EventSystem conventions | requiredRootsJson?, requiredTagsJson?, requiredLayersJson?, requireEventSystemForUi? |
High-Frequency Skill Differences
scene_summarize vs scene_analyze vs scene_health_check
| Skill | Best for | Typical output focus |
scene_summarize | Fast overview | object counts, hierarchy depth, top components |
scene_analyze | Broad diagnosis | summary + findings + warnings + recommendations + next-skill hints |
scene_health_check | Hygiene / red flags | missing scripts, duplicate names, deep hierarchy, empty nodes, hotspot-style findings |
scene_context vs hierarchy_describe
| Skill | Best for | Output style |
hierarchy_describe | Human-readable tree | text tree, lightweight |
scene_context | AI coding context | structured hierarchy + components + references + optional code dependencies |
scene_dependency_analyze vs script_dependency_graph
| Skill | Scope | Use when |
scene_dependency_analyze | Scene object references | ask "删掉/禁用这个对象会影响谁" |
script_dependency_graph | Script class dependency closure | ask "要改这个功能先看哪些脚本" |
Key Return Shapes
scene_summarize
Returns sceneName, stats, and optional topComponents.
scene_analyze
Returns summary, stats, findings, warnings, recommendations, and suggestedNextSkills.
scene_health_check
Returns summary, findings, hotspots, and suggestedNextSkills.
scene_context
Returns a structured export with objects, references, and optional codeDependencies. Use it when another AI step needs full scene context, not just a human summary.
High-frequency options:
rootPathto export only one subtreeincludeValues=truewhen serialized field values matterincludeCodeDeps=truewhen AI needs a rough scene-to-code dependency picture
scene_export_report
Writes a markdown artifact to disk and returns savedTo, object/script/reference counts, and success state. Prefer this when the user wants a durable report file.
Defaults:
savePath = "Assets/Docs/SceneReport.md"maxDepth = 10maxObjects = 500
When to Use Which Skill
| Need | Best first skill |
| Quick scene overview | scene_summarize |
| Full diagnosis | scene_analyze |
| Suspicious hierarchy or clutter | scene_find_hotspots |
| Safe-to-delete / impact question | scene_dependency_analyze |
| AI coding context export | scene_context |
| Script reading order | script_dependency_graph |
| Render/input/UI stack detection | project_stack_detect |
RunCommand Examples
Recipe path rule: ../../recipes/perception/<command>.md
See ../../recipes/perception/README.md for the full command index and per-command C# templates.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.