scientific-prose-style

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill consists entirely of natural language instructions for text formatting and prose style. It provides rules for em-dash usage, sentence length, and paragraph construction in scientific writing.
  • [NO_CODE]: There are no scripts, command-line executions, network operations, or external dependencies. The skill operates solely as a prompt-based guide for stylistic editing.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided manuscript text, which is an ingestion point for untrusted data. However, the capabilities are restricted to stylistic analysis and rewriting instructions with no access to sensitive system tools, file-writing, or network operations, posing no significant security risk.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 12:00 PM
Security Audit — agent-trust-hub — scientific-prose-style