file-intel

Fail

Audited by Gen Agent Trust Hub on Mar 16, 2026

Risk Level: HIGHCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to run python scripts/process_files_with_gemini.py <folder_path> where the path is provided by the user. There are no instructions for input validation or shell escaping, which allows an attacker to execute arbitrary system commands by providing a folder path containing shell metacharacters (e.g., ;, |, or &&).
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it extracts and summarizes content from various external file formats (PDF, DOCX, CSV, Python, etc.).
  • Ingestion points: Files within the user-specified folder are read and processed (SKILL.md).
  • Boundary markers: Absent; the instructions do not specify any delimiters or safety prompts to prevent the LLM from following instructions embedded within the files being processed.
  • Capability inventory: The skill has the ability to execute shell commands and open local file paths.
  • Sanitization: No sanitization of the file content is described before the content is passed to the Gemini processor.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Mar 16, 2026, 06:36 AM