mobile-security-mobsf
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to download the official MobSF Docker image from the
opensecurityorganization to perform analysis.- [COMMAND_EXECUTION]: The instructions include templatecurlcommands to interact with the MobSF REST API on a local host for uploading binaries, triggering scans, and retrieving reports.- [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection due to its core function of processing external application data. - Ingestion points: Untrusted mobile application binaries (APK, IPA, ZIP) and the resulting JSON scan reports are processed by the agent to generate summaries.
- Boundary markers: The instructions lack explicit delimiters or instructions to treat data from the scan reports as untrusted content, which could lead the agent to follow instructions embedded within the analyzed application's metadata or code.
- Capability inventory: The skill utilizes
curlfor network communication with a local service and performs file system writes by redirecting command output to JSON files. - Sanitization: No sanitization or validation logic is defined for the data extracted from the external scan reports before it is presented to the user.
Audit Metadata