writing-beats

Pass

Audited by Gen Agent Trust Hub on Sep 20, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided markdown files and re-reads article content from the disk, creating a surface where instructions embedded in the raw text could potentially influence agent behavior.
  • Ingestion points: The skill instructions specify reading from a 'markdown file of raw material' and an 'article file from disk'.
  • Boundary markers: There are no explicit delimiters or instructions provided to the agent to disregard potential instructions contained within the ingested material.
  • Capability inventory: The skill is granted the ability to read from and write to the local file system at paths defined by the user.
  • Sanitization: The instructions do not mention any sanitization or validation of the text retrieved from external files.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 20, 2026, 07:48 AM
Security Audit — agent-trust-hub — writing-beats