Plan March Break

Pass

Audited by Gen Agent Trust Hub on Jun 30, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes data from untrusted external sources.
  • Ingestion points: The agent is instructed to perform web searches for camp programs and download school calendar PDFs provided by the user or found online (SKILL.md).
  • Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from following commands that might be embedded within those external PDFs or websites.
  • Capability inventory: The skill has the capability to write to the local filesystem (creating schedule and budget files), perform web searches, download files, and execute other internal scripts/skills.
  • Sanitization: No validation or sanitization steps are specified for the content of the external files or web search results before the agent processes them.
  • [EXTERNAL_DOWNLOADS]: The workflow requires the agent to search for and download school calendars and camp program details from various websites to extract scheduling data.
  • [COMMAND_EXECUTION]: The skill facilitates the execution of internal scripts and tools, such as a budget calculator and the 'generate-annual-schedule' skill, to finalize the planning data.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 30, 2026, 11:31 AM
Security Audit — agent-trust-hub — Plan March Break