extracting-with-ocr
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PRIVILEGE_ESCALATION]: The skill instructs users or agents to use
sudo apt installto add Tesseract language packs, which involves running commands with root privileges. - [INDIRECT_PROMPT_INJECTION]: The skill enables the extraction of text from potentially untrusted external files (images and PDFs), creating a vector for malicious instructions to enter the agent's context.
- Ingestion points: File text extraction using the
xberg extractcommand as documented in SKILL.md. - Boundary markers: The skill does not provide instructions for the use of delimiters or warnings to ignore instructions found within the processed files.
- Capability inventory: The skill relies on the
xbergCLI tool to perform file access and text processing. - Sanitization: No sanitization or verification steps for the extracted content are mentioned in the skill documentation.
- [COMMAND_EXECUTION]: The documentation details multiple CLI operations using the
xbergtool and system package managers (brew,apt) to perform text extraction and environment setup.
Audit Metadata