oceanbase-formatting

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [Prompt Injection] (SAFE): The skill uses instructional language to enforce formatting standards without attempting to bypass safety filters or override system-level instructions.
  • [Data Exposure & Exfiltration] (SAFE): No sensitive file paths, credentials, or data exfiltration patterns were detected.
  • [Obfuscation] (SAFE): No encoded or hidden content, such as Base64 or zero-width characters, was found.
  • [External Downloads] (SAFE): References a static image URL for a download icon, which is used for documentation display purposes only.
  • [Indirect Prompt Injection] (SAFE): The skill processes user documentation (Ingestion point: User-provided text in SKILL.md). It lacks explicit boundary markers or sanitization, but since it has no dangerous capabilities like shell execution or network requests (Capability inventory: None), the risk of exploitation is negligible.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:30 PM