scene-feedback
Skill4 starsChanged 10 days agoApache-2.0View on GitHub
Scene Feedback surveys for Decentraland SDK7 playtests: plan questions and timing, organize test rounds and find testers, integrate or debug the module, or analyze exported CSV ratings and comments. For audiovisual reactions to gameplay actions, use add-interactivity, audio-video or build-ui.
In this repository37 other files it gives its agents.
- add-3d-modelsSkill
- add-interactivitySkill
- advanced-inputSkill
- advanced-renderingSkill
- animations-tweensSkill
- audio-analysisSkill
- audio-videoSkill
- authoritative-serverSkill
- avatar-animationsSkill
- build-uiSkill
- camera-controlSkill
- compositesSkill
- create-sceneSkill
- creator-hub-mcpSkill
- dcl-gddSkill
- dcl-prototypeSkill
- deploy-sceneSkill
- deploy-worldsSkill
- fontsSkill
- lighting-environmentSkill
- migrate-sdk6-to-sdk7Skill
- multiplayer-syncSkill
- nft-blockchainSkill
- npcsSkill
- optimize-sceneSkill
- particle-systemSkill
- player-avatarSkill
- player-physicsSkill
- promote-sceneSkill
- report-sdk-issueSkill
- scene-retroSkill
- scene-runtimeSkill
- script-componentsSkill
- sdk-scenesSkill
- smart-wearablesSkill
- unity-explorer-mcpSkill
- wearablesSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
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 31
When updating project notes, distinguish creator-approved decisions, assistant proposals, and implemented facts.
- Distinctive: line 36
Use ordinary language matched to how familiar the creator is with the tools, and explain a technical term through its purpose the first …
≈ “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
Scene Feedback
Help creators learn how players experience their scene using Scene Feedback questions and CSV responses.
REQUIRED gate: approved plan
Approved plan gate: before writing receiving-scene files for a new integration, or making content or timing edits the creator did not dictate, the creator has explicitly approved each of:
- goal,
- questions and Intro,
- formats and scales,
- moments,
- participation route.
An assistant suggestion, including one from an earlier session, stays a proposal until the creator confirms it. A request to "install it" or "connect it", or approval of a technical option, is not approval of these choices. An approved exact edit or plan proceeds without a renewed interview.
Any item missing → read Planning and Scene context, then present a compact plan covering the missing items, with the three planning paths. The gate passes when the creator accepts that concrete plan.
When updating project notes, distinguish creator-approved decisions, assistant proposals, and implemented facts; implementation itself does not establish approval.
Creator-facing review
Use ordinary language matched to how familiar the creator is with the tools, and explain a technical term through its purpose the first time it appears (an invitation asking players whether to give feedback: the Intro). Show question wording, rating label meaning and player moments, with any bank, scale or event ID beside its meaning. Keep exact IDs in code and configuration.
Choose the route
| Request | Read next | Scope |
| Choose a playtest focus, propose/adapt questions, scales, or timing | Planning | Ground the proposal in the scene or brief and review creator-facing choices before applying them. |
| Organize test rounds: who to invite, round cadence, live or asynchronous, finding testers | Run test rounds | Advice only; writes no scene files. |
| Install, wire events, apply approved edits, or update the module | Integration | Inspect the current installation first; reuse it and change only what the task needs. |
| Check/debug existing behavior or test devices/Storage | Verification | Inspect configuration and hooks, then run relevant checks. |
| Download responses only | World Storage retrieval | Explain retrieval without starting a full analysis. |
| Analyze collected CSV ratings/comments or compare playtests | Analysis | Start with the exports and goal; offer optional HTML. |
Enter at the matching row and stop at its scope. Reuse supplied goals and approved decisions. For an existing integration, distinguish a question/configuration edit from installation or update; for CSV analysis, the exports can be sufficient without access to scene source. The module docs linked from each workflow are the source of truth for configuration, API, examples and CSV semantics; disclose any version mismatch between them and the scene that affects the work.
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.