box
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of external data from Box (files, folders, and search results). This creates a potential surface for indirect prompt injection if the agent processes untrusted content retrieved from storage.
- Ingestion points: Data enters the context via
apideck.fileStorage.files.listand file retrieval methods mentioned inSKILL.md. - Boundary markers: No explicit delimiters or instructions to ignore embedded commands are specified in this connector skill.
- Capability inventory: The skill enables file creation, writing, and direct API interaction via a proxy service in
SKILL.md. - Sanitization: No sanitization or content filtering mechanisms are described for the retrieved data.
- [EXTERNAL_DOWNLOADS]: The skill references the
@apideck/unifyNode.js package. This is a vendor-owned resource used for interacting with the Apideck unified API platform. - [DATA_EXPOSURE]: The skill provides guidance on using environment variables (
process.env.APIDECK_API_KEY) for API key management, which aligns with security best practices for avoiding hardcoded credentials.
Audit Metadata