window-scenery
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to run a bundled Python script (scripts/overlay_badge.py) to add location markers to images. This involves executing shell commands with arguments.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user input and uses it as arguments for command-line execution, creating a vulnerability surface.
- Ingestion points: User-provided strings for the location (place) and coordinates (coords) extracted from requests in SKILL.md.
- Boundary markers: None; the skill does not define clear boundaries or instructions to prevent the agent from interpreting instructions embedded within the user-provided location data.
- Capability inventory: The agent is authorized to execute a local Python script via the shell, as specified in the Location Badge section of SKILL.md.
- Sanitization: There is no instruction to validate or escape user input before it is passed to the python command, relying solely on the agent's internal handling of shell arguments.
Audit Metadata