triage

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from the user's Readwise inbox to generate summaries and 'pitches'.
  • Ingestion points: Document content and summaries are fetched via mcp__readwise__reader_list_documents and mcp__readwise__reader_get_document_details in SKILL.md.
  • Boundary markers: The instructions lack explicit boundary markers or warnings to the agent to ignore instructions embedded within the fetched document text.
  • Capability inventory: The agent can perform state-changing actions such as moving or archiving documents using mcp__readwise__reader_move_document or CLI equivalents.
  • Sanitization: There is no evidence of sanitization or filtering applied to the external document data before it is processed by the LLM.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 03:09 PM