avoid-numeric-index

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOWNO_CODE
Full Analysis
  • [NO_CODE] (SAFE): The skill primarily consists of markdown documentation and TypeScript code examples for educational purposes. It does not implement any logic that executes system commands or handles external data.
  • [PROMPT_INJECTION] (SAFE): Analysis of the markdown content found no attempts to override agent instructions, bypass safety filters, or extract system prompts. The 'Iron Rule' section is a stylistic guideline for code development, not an instruction for the AI agent to ignore its core safety parameters.
  • [DATA_EXFILTRATION] (SAFE): No network requests, file system access, or credential markers were detected in the provided code snippets.
  • [INDIRECT_PROMPT_INJECTION] (SAFE): This skill does not ingest untrusted external data and has no capability to perform write operations or external communication, presenting no surface for injection attacks.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 01:10 PM