docx-reader
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- PROMPT_INJECTION (LOW): The skill is susceptible to Indirect Prompt Injection. It extracts text from external .docx files and provides it to the agent without using boundary markers or sanitization. If a document contains hidden instructions (e.g., 'ignore previous instructions'), the agent may follow them.
- Ingestion points: text output from
scripts/read_docx.py. - Boundary markers: Absent. The agent is not instructed to treat the extracted text as untrusted data.
- Capability inventory: The agent can execute shell commands (
wsl) and read/write files. - Sanitization: Absent. The script extracts raw text and table data directly.
- COMMAND_EXECUTION (LOW): The skill requires the agent to execute a local Python script via the command line. While necessary for functionality, this pattern involves the agent constructing and running shell commands based on user-provided file paths.
Audit Metadata