study-notes-generator
Pass
Audited by Gen Agent Trust Hub on Feb 21, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- COMMAND_EXECUTION (LOW): The skill instructs the agent to execute a specific local binary (
/opt/homebrew/bin/pdftotext) to process large PDF files. While this is the intended primary purpose of the skill, any shell command execution involves a potential risk if inputs (like filenames) are not handled securely.\n- PROMPT_INJECTION (LOW): The skill processes untrusted documents from theslides/directory and is vulnerable to indirect prompt injection.\n - Ingestion points: PDF and TXT documents in the
slides/folder.\n - Boundary markers: Absent. There are no instructions for the agent to use delimiters or specific logic to ignore instructions embedded within the processed text.\n
- Capability inventory: Shell command execution (
pdftotext), file-system write access to thenotes/directory, and the ability to spawn background agents via theTasktool.\n - Sanitization: Absent. The agent is not instructed to sanitize or escape HTML tags found in the source documents before placing them into the HTML template, which could lead to XSS if a user opens the generated files in a browser.
Audit Metadata