NYC

search-strategy

Pass

Audited by Gen Agent Trust Hub on Feb 15, 2026

Risk Level: LOWPROMPT_INJECTION
Full Analysis
  • PROMPT_INJECTION (LOW): Indirect Prompt Injection Surface. The instructions guide the agent to decompose queries and synthesize results from external sources like chat and wiki.
  • Ingestion points: Data enters the context from '~~chat', '~~knowledge base', and '~~project tracker' searches.
  • Boundary markers: The skill lacks instructions for using delimiters or protective prompts to isolate retrieved content.
  • Capability inventory: Synthesis and search translation are the primary capabilities; no dangerous write or execute operations are defined within this skill file.
  • Sanitization: No logic is provided to sanitize or escape potentially malicious instructions embedded in search results.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 15, 2026, 08:29 PM