deep-reader

Pass

Audited by Gen Agent Trust Hub on Apr 26, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious behavior, unauthorized command execution, or sensitive data access patterns were identified. The skill's functionality is strictly limited to processing and analyzing user-provided content.\n- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it is designed to ingest and process untrusted external data.\n
  • Ingestion points: The skill reads local files (PDF, DOCX, TXT, MD, EPUB) via a read tool and fetches external content from URLs using a WebFetch tool.\n
  • Boundary markers: Absent. There are no explicit instructions for the agent to treat the fetched content as untrusted data or to disregard any instructions contained within it.\n
  • Capability inventory: The skill uses file reading and web fetching capabilities to provide input for its analysis commands.\n
  • Sanitization: Not specified in the skill's instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 26, 2026, 06:18 PM