mobilerun-assistant
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill utilizes official software development kits and interacts with API endpoints belonging to the vendor (droidrun).\n
- Node.js Package: '@mobilerun/sdk' (SKILL.md)\n
- Python Package: 'mobilerun-sdk' (SKILL.md)\n
- API Endpoint: 'https://api.mobilerun.ai/v1' (SKILL.md)\n- [COMMAND_EXECUTION]: The skill provides examples using 'curl' and 'jq' to interact with the Mobilerun assistant's REST API, which is standard for its intended use case.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes data from an external assistant service, creating a surface for indirect prompt injection.\n
- Ingestion points: Assistant text deltas and tool call responses from the SSE stream and session history in 'SKILL.md'.\n
- Boundary markers: The skill does not demonstrate the use of specific boundary markers or instructions to ignore embedded commands when processing assistant output.\n
- Capability inventory: The skill allows the agent to approve sensitive permissions ('answerPermission'), answer clarifying questions ('answerQuestion'), and manage chat sessions ('SKILL.md').\n
- Sanitization: No explicit sanitization of the streamed content is described in the documentation.\n- [SAFE]: No malicious patterns, obfuscated code, or unauthorized data access behaviors were detected. The skill's implementation is consistent with its stated purpose.
Audit Metadata