drive9
Pass
Audited by Gen Agent Trust Hub on Apr 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: Executes shell commands using
curlto interact with the drive9.ai REST API for file and directory management (GET, PUT, POST, DELETE). - [DATA_EXFILTRATION]: Transmits content from the local environment (e.g., files in
/tmp/) to the externalapi.drive9.aiendpoint. This behavior is expected given the skill's purpose as a network filesystem. - [PROMPT_INJECTION]: The skill facilitates the ingestion and retrieval of persistent data, which creates a surface for indirect prompt injection.
- Ingestion points: Untrusted data enters the context through
curl -X PUToperations and directory listing described inSKILL.md. - Boundary markers: Absent. The instructions do not define delimiters or provide guidance to ignore instructions within stored files.
- Capability inventory: The skill uses
curlfor network requests and can write/delete files via the API, as seen inSKILL.md. - Sanitization: No sanitization or validation of the stored file content is mentioned or implemented in the provided instructions.
Audit Metadata