pdf_translator

Warn

Audited by Gen Agent Trust Hub on May 11, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands that directly incorporate user-controlled input (e.g., python3 skills/pdf_translator/scripts/extract_text.py <path_to_pdf>). This creates a vulnerability where a malicious filename containing shell metacharacters (such as ;, &, or |) could be used to execute arbitrary commands on the host system.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it extracts text from untrusted PDF documents and presents it to the agent for translation without isolation.
  • Ingestion points: Text extracted from PDFs using the scripts/extract_text.py script (referenced in SKILL.md).
  • Boundary markers: Absent. There are no delimiters or instructions provided to the agent to treat the extracted PDF content as untrusted data.
  • Capability inventory: The skill has the ability to execute local Python scripts and perform file write operations via scripts/generate_md.py.
  • Sanitization: Absent. The extracted content is passed directly into the agent's context for processing without any filtering or escaping of potential instructions.
Audit Metadata
Risk Level
MEDIUM
Analyzed
May 11, 2026, 07:10 AM
Security Audit — agent-trust-hub — pdf_translator