ljg-rank
Pass
Audited by Gen Agent Trust Hub on Mar 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute the shell command
dateto generate formatted timestamps for use in filenames and document metadata. - [PROMPT_INJECTION]: The skill demonstrates an indirect prompt injection surface by interpolating user-provided input ('领域' or domain) into a local file system path and shell-derived filename without providing sanitization or validation logic.
- Ingestion points: User input provided for the '领域' variable in the prompt.
- Boundary markers: Absent; user-provided data is used directly to construct the file path and name.
- Capability inventory: Local file system write access (
~/Documents/notes/) and shell command execution (date). - Sanitization: Absent; there are no instructions to escape characters or validate the input before it is used in the file path construction.
Audit Metadata