decentraland-sdk-skills
Pass
Audited by Gen Agent Trust Hub on May 8, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as an authoritative context provider for Decentraland SDK7 development, authored by the official vendor. It focuses on teaching the 'Composite-first' architecture and correct SDK usage.
- [EXTERNAL_DOWNLOADS]: The skill references downloads for 3D models and audio assets from official Decentraland domains (e.g.,
builder-items.decentraland.org,models.dclregenesislabs.xyz). These are documented neutrally as part of the asset discovery workflow. - [COMMAND_EXECUTION]: Documented commands (e.g.,
npx @dcl/sdk-commands init,npm install,curl) are standard tools for scaffolding and managing Decentraland projects. The.claude/settings.jsonfile provides specific, limited permissions for the agent to fetch documentation updates from the author's GitHub repository. - [DATA_EXFILTRATION]: Networking capabilities like
fetch,signedFetch, andWebSocketare presented as core features for multiplayer synchronization and backend connectivity. The skill actively promotes secure practices, such as usingsignedFetchto verify player identity and using environment variables for sensitive keys.
Audit Metadata