text-formatter
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [Prompt Injection] (SAFE): No instructions attempting to bypass safety filters or override agent behavior were detected in the skill definitions.
- [Data Exposure & Exfiltration] (SAFE): The skill manages local files for text processing but does not perform network operations or access sensitive system directories like SSH keys or environment variables.
- [Remote Code Execution] (SAFE): There are no patterns involving the download or execution of remote scripts, nor any use of package managers like npm or pip.
- [Indirect Prompt Injection] (LOW): The skill processes untrusted user-provided text, which is an inherent surface for indirect prompt injection. However, the capabilities are limited to formatting tasks.
- Ingestion points: Reads user-provided text or files in Phase 1.
- Boundary markers: None explicitly defined in the top-level SKILL.md.
- Capability inventory: Includes Read, Write, and Bash (used for directory creation).
- Sanitization: No explicit sanitization or escaping of external content is mentioned.
- [Dynamic Execution] (SAFE): The Bash tool is used only for creating a timestamped scratchpad directory (mkdir -p), which is a standard and low-risk operation.
Audit Metadata