privacy-mask
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires the installation of the
privacy-maskpackage viapipand system-level OCR tools (tesseract). These dependencies originate from community-maintained sources rather than established trusted vendors. - [COMMAND_EXECUTION]: The skill instructions direct the agent to execute the
privacy-maskCLI tool with various flags, including--in-placeand--config, which involve local file system operations and shell command execution. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted image data (screenshots and photos) which may contain embedded instructions designed to influence the agent's behavior during the interpretation of the tool's output.
- Ingestion points: Reads image files (e.g.,
.png,.jpg,.bmp) from local paths provided by the user. - Boundary markers: None present to separate image content from agent instructions.
- Capability inventory: Executes shell commands via the
privacy-maskCLI tool. - Sanitization: There is no evidence of sanitization or filtering of the text extracted from images before the agent processes the resulting JSON output.
Audit Metadata