writing-shape

Pass

Audited by Gen Agent Trust Hub on Jul 23, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill operates on untrusted external markdown files and has file system access, which presents a surface for indirect prompt injection.\n
  • Ingestion points: The skill reads a 'raw material' markdown file provided by the user as its primary data source (SKILL.md).\n
  • Boundary markers: There are no instructions to use specific delimiters (like XML tags or unique markers) to separate the untrusted input text from the system instructions.\n
  • Capability inventory: The skill includes instructions to read from the disk and append/write paragraphs to an article file (SKILL.md).\n
  • Sanitization: The skill lacks logic to sanitize or filter the content of the input pile for embedded malicious instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 23, 2026, 11:19 AM
Security Audit — agent-trust-hub — writing-shape