skills/team-attention/hoyeon/stepback/Gen Agent Trust Hub

stepback

Pass

Audited by Gen Agent Trust Hub on Apr 8, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill ingests untrusted data from conversation history and local files to generate its findings, which creates a surface for indirect prompt injection.
  • Ingestion points: Conversation history and file system via Read, Grep, and Glob tools.
  • Boundary markers: No delimiters or isolation instructions are defined for the processed data.
  • Capability inventory: Tools are limited to data retrieval; code execution and agent recursion are strictly forbidden.
  • Sanitization: No input validation or sanitization is mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 8, 2026, 11:01 AM