regex-builder

Pass

Audited by Gen Agent Trust Hub on Apr 11, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill facilitates an indirect prompt injection surface by instructing the agent to ingest untrusted user data to generate complex regex patterns and code snippets.
  • Ingestion points: The workflow in SKILL.md (Phase 1) explicitly collects positive and negative examples from the user and context to drive pattern generation.
  • Boundary markers: The instructions do not define boundary markers or 'ignore' directives to prevent the agent from obeying instructions that might be embedded within the provided string examples.
  • Capability inventory: The skill generates functional Python and JavaScript code templates (SKILL.md) and regex patterns based on the untrusted examples.
  • Sanitization: The skill lacks mechanisms for sanitizing or validating user-provided examples before they are interpolated into the pattern inference and code generation process.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 11, 2026, 01:24 PM
Security Audit — agent-trust-hub — regex-builder