Camp Planner Setup

Warn

Audited by Gen Agent Trust Hub on Jun 30, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill allows the user to define a custom name for the research directory, which is subsequently used to construct paths for file read and write operations. The absence of input validation for this name introduces a risk of path traversal, potentially allowing the agent to access or modify files outside the intended project directory.
  • [EXTERNAL_DOWNLOADS]: The skill automates the discovery and downloading of school calendar PDF files from various external websites based on user-provided school board information.
  • [PROMPT_INJECTION]: The skill ingests untrusted data from downloaded PDFs, creating a surface for indirect prompt injection.\n
  • Ingestion points: External PDF files and web search results for school calendars.\n
  • Boundary markers: The instructions provide no delimiters or directives to ignore instructions that may be embedded within the external PDF content.\n
  • Capability inventory: The skill uses tools for reading, writing, and searching the web to process and store external data.\n
  • Sanitization: There is no mention of sanitizing or validating the text extracted from the external PDFs before it is saved to the library or used in the agent's context.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 30, 2026, 11:31 AM
Security Audit — agent-trust-hub — Camp Planner Setup