wjs-distilling-style

Pass

Audited by Gen Agent Trust Hub on Jun 21, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides example curl commands for users to interact with the jianshuo.dev API. These commands are intended for managing writing style profiles through listing, downloading, and uploading operations.
  • [EXTERNAL_DOWNLOADS]: The skill references the vendor domain jianshuo.dev for optional profile synchronization. This is an established domain owned by the author and used for legitimate tool functionality.
  • [PROMPT_INJECTION]: The skill analyzes user-provided text samples to extract stylistic fingerprints. Although processing external data presents a theoretical surface for indirect prompt injection, the risk is mitigated by the skill's structured analysis approach and the use of isolated sub-agents for evaluation tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 21, 2026, 05:46 AM
Security Audit — agent-trust-hub — wjs-distilling-style