tutor-setup

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute shell commands for processing files, specifically using pdftotext to convert PDF documents into text format and using system package managers to resolve missing dependencies.
  • [EXTERNAL_DOWNLOADS]: The instructions direct the agent to install the poppler utility using official package managers (brew for macOS or apt-get for Linux) if the tool is missing from the host environment.
  • [INDIRECT_PROMPT_INJECTION]: The skill's primary function involves ingesting and summarizing untrusted external data, including PDF files, web content via WebFetch, and source code.
  • Ingestion points: Phase D1 (PDF/text/web discovery) and Phase C1 (Codebase exploration).
  • Boundary markers: The skill lacks explicit delimiters or instructions to ignore embedded prompts within the source material.
  • Capability inventory: The agent has access to Bash, Write, and WebFetch tools, which could be leveraged if malicious instructions were successfully injected through source documents.
  • Sanitization: No specific content sanitization or validation steps are defined for the data extracted from source files before it is processed by the LLM.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 03:22 PM
Security Audit — agent-trust-hub — tutor-setup