weaverse-content-api
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides a Node.js script
scripts/weaverse_content_api.mjswhich the agent is instructed to execute using thenodecommand. The script performs local file system reads and network operations to synchronize content. - [DATA_EXFILTRATION]: The script
scripts/weaverse_content_api.mjsreads environment variables (WEAVERSE_API_KEY) and local files (e.g.,patch.json,theme.json, and media files) and transmits them to vendor-owned endpoints atstudio.weaverse.ioandapi.weaverse.io. These operations are consistent with the skill's primary function of content synchronization and media management. - [INDIRECT_PROMPT_INJECTION]: The skill's architecture involves reading live website content from an external API, which creates a potential surface for indirect prompt injection if the project data contains malicious instructions.
- Ingestion points: Content is retrieved via
GETrequests to the Weaverse Content API, as documented inreferences/endpoints.mdand implemented inscripts/weaverse_content_api.mjs. - Boundary markers: The instructions do not define specific delimiters or security warnings to prevent the agent from following instructions potentially embedded in the retrieved content.
- Capability inventory: The skill possesses broad write permissions, including the ability to
PATCHpage content,POSTnew pages,DELETEproject resources, and upload files to Shopify through theadmin-graphqlproxy inscripts/weaverse_content_api.mjs. - Sanitization: There is no mention of sanitization, filtering, or validation of the data retrieved from the Weaverse project before it is processed by the agent.
Audit Metadata