task-automation
Warn
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: MEDIUMPERSISTENCEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PERSISTENCE]: The skill provides instructions and examples for creating recurring automated tasks using cron jobs and systemd timers, which function as persistence mechanisms to ensure code execution across system restarts and sessions.
- [INDIRECT_PROMPT_INJECTION]: The skill contains a vulnerability surface where untrusted data could influence agent behavior.
- Ingestion points: The implementation examples demonstrate processing external CSV files from the filesystem (e.g.,
~/data/incoming/). - Boundary markers: The provided instructions and scripts lack explicit delimiters or instructions for the agent to ignore natural language commands that might be embedded within the processed data files.
- Capability inventory: The skill possesses capabilities for file system manipulation, including reading files, writing new data (
json.dump), renaming files (filepath.rename), and deleting files (filepath.unlink). - Sanitization: Although the example script validates CSV headers, it does not sanitize or escape the content of the data rows before processing or transforming them, creating a risk that malicious content in the data could be interpreted as instructions by downstream components.
- [COMMAND_EXECUTION]: The skill's primary workflow involves designing, writing, and executing shell scripts and command-line tools to automate repetitive manual tasks.
Audit Metadata