professional-proofreader

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process text from external document files, which can be used to deliver hidden instructions to the agent.\n
  • Ingestion points: As specified in references/file-processing-mode.md, the skill reads content from .docx, .txt, and .pdf files.\n
  • Boundary markers: The instructions do not define the use of delimiters or specific prompts to ensure the agent ignores any instructions embedded within the processed text.\n
  • Capability inventory: The skill instructions in SKILL.md and references/file-processing-mode.md permit both reading from the file system and writing new files.\n
  • Sanitization: There is no mention of sanitizing, escaping, or validating the text extracted from documents to prevent the execution of embedded prompts.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 07:11 PM
Security Audit — agent-trust-hub — professional-proofreader