location-street-view-satellite-imagery
Pass
Audited by Gen Agent Trust Hub on Jun 23, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes external, untrusted input (addresses) which creates a surface for indirect prompt injection attacks. * Ingestion points: The 'address' parameter across multiple actions including 'geocode', 'get_street_view_image', and 'get_satellite_image'. * Boundary markers: No delimiters or explicit 'ignore embedded instructions' warnings are provided to the agent for these inputs. * Capability inventory: The skill allows the agent to call remote geocoding and imagery services, which return external data and signed URLs. * Sanitization: There are no specific instructions for the agent to validate or sanitize the input strings before they are passed to the tool actions.- [EXTERNAL_DOWNLOADS]: Mentions the vendor's official GitHub repository (AgentPMT/agent-skills) as the source for skill updates and installation, which is a trusted platform resource.- [COMMAND_EXECUTION]: Provides standard documentation for user-led skill installation and maintenance using the npx package runner.
Audit Metadata