crypto-info-archive

Pass

Audited by Gen Agent Trust Hub on Mar 18, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill instructions facilitate the processing of untrusted external content, presenting an indirect prompt injection surface.
  • Ingestion points: Technical documentation in references/sentiment-data.md describes processing news headlines and full-text content from external market sources.
  • Boundary markers: The skill lacks explicit instructions for the agent to utilize boundary markers or delimiters when handling external data to prevent command obedience within that data.
  • Capability inventory: The skill involves network operations (HTTP POST/GET) to transmit data to the 'aitrade' support server, potentially including malicious instructions from ingested news.
  • Sanitization: There is no mention of content sanitization, filtering, or validation before formatting and archiving the external data.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 18, 2026, 09:03 PM