feed-catchup

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [SAFE]: All external resources and tools identified are official components of the Readwise ecosystem, consistent with the skill's authorship by readwiseio.
  • [COMMAND_EXECUTION]: The skill utilizes the 'readwise' CLI for data management if MCP tools are unavailable. These commands (e.g., 'readwise list', 'readwise move') are vendor-specific and used for intended functionality.
  • [PROMPT_INJECTION]: Evaluated for indirect prompt injection risk as the skill processes content from external RSS feeds.
  • Ingestion points: Document metadata and full content retrieved via mcp__readwise__reader_list_documents and mcp__readwise__reader_get_document_details (SKILL.md).
  • Boundary markers: External content is presented in Markdown tables and summaries without specific instruction-isolation delimiters.
  • Capability inventory: Limited to Readwise-specific operations: listing documents, reading content, and updating metadata or location within the user's account (SKILL.md).
  • Sanitization: Standard processing of text metadata for display and summarization.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 05:46 PM