source-management
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE] (SAFE): The file SKILL.md is a documentation and prompt-engineering file. It provides logic for prioritizing search results and handling rate limits but contains no Python, JavaScript, or shell scripts that could execute on the host system.
- [PROMPT_INJECTION] (LOW): The skill outlines a system for ingesting data from untrusted external sources (email, chat, CRM), creating a surface for indirect prompt injection. Ingestion points: Search tool outputs (chat, email, docs). Boundary markers: Not specified in the provided text. Capability inventory: Search and retrieval tools. Sanitization: Not mentioned in the instructions.
Audit Metadata