summarize

Pass

Audited by Gen Agent Trust Hub on May 9, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local Python script scripts/extract_pdf_pages.py to extract text from PDF files. While intended for processing large documents, it represents a local execution vector.
  • [EXTERNAL_DOWNLOADS]: Users are instructed to install the pymupdf library via pip to enable PDF text extraction functionality.
  • [DATA_EXFILTRATION]: The workflow specifies that summaries should be uploaded to Notion automatically using MCP tools without seeking user confirmation. This autonomy increases the risk of unintended data flow to external services.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted document content (PDF, EPUB). Ingestion points: Content is read from user-uploaded documents or pasted text in Step 1. Boundary markers: There are no explicit delimiters or instructions to ignore embedded commands in the provided files. Capability inventory: The skill can write local .md files, execute Python scripts, and perform Notion workspace operations (search and page creation). Sanitization: There is no evidence of input sanitization or validation of the document content before it is processed or uploaded.
Audit Metadata
Risk Level
SAFE
Analyzed
May 9, 2026, 02:00 PM