regex-whisperer
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided regular expressions and text, creating a potential surface for indirect prompt injection attacks where malicious patterns could influence agent behavior.
- Ingestion points: The skill description in SKILL.md indicates it triggers on regex-related keywords and processes patterns or text for validation and debugging.
- Boundary markers: The skill instructions emphasize grounding responses in provided reference files, which serves as a behavioral constraint.
- Capability inventory: The skill does not possess capabilities for network communication, sensitive file access, or shell command execution.
- Sanitization: Extensive reference material (references/sharp_edges.md and references/validations.md) is provided to identify and mitigate dangerous input patterns, such as Regular Expression Denial of Service (ReDoS).
Audit Metadata