writing-beats

Pass

Audited by Gen Agent Trust Hub on Jul 8, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill involves reading from and writing to the local file system. It reads a user-provided markdown file of "raw material" and writes the generated article to a destination path specified by the user. It also re-reads the article from disk during the process to ensure user edits are preserved.
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it processes untrusted content from an external markdown file provided by the user.
  • Ingestion points: The agent reads content from a markdown file specified by the user as "raw material."
  • Boundary markers: Absent. There are no instructions to use delimiters or ignore embedded commands within the processed file.
  • Capability inventory: The skill can read and write files on the local disk based on user-provided paths.
  • Sanitization: Absent. No explicit sanitization or filtering of the source material is mentioned before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 8, 2026, 01:57 AM
Security Audit — agent-trust-hub — writing-beats