nano-pdf
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructions direct the agent or user to install the
nano-pdfpackage from the Python Package Index (PyPI), which is a well-known public registry. \n- [COMMAND_EXECUTION]: The skill uses thenano-pdfcommand-line tool to perform file editing operations, which involves executing shell commands with user-supplied arguments. \n- [INDIRECT_PROMPT_INJECTION]: The skill processes PDF content and natural language instructions through an underlying LLM-powered tool. This creates a vulnerability surface where adversarial content within a PDF file could attempt to hijack the tool's logic or the agent's context. \n - Ingestion points: Processes PDF file contents (
file.pdf) and instruction strings provided in the command line. \n - Boundary markers: The skill does not define specific delimiters or instructions to prevent the LLM from following commands embedded inside the PDF files being edited. \n
- Capability inventory: The skill possesses the capability to execute shell commands (
nano-pdf edit ...) and read/modify files. \n - Sanitization: The documentation does not specify any sanitization or filtering of the PDF content before it is processed by the LLM.
Audit Metadata