cron-tuner
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill consists entirely of markdown documentation and instructions. It does not contain any executable scripts (Python, JavaScript, Shell) or binaries.
- [COMMAND_EXECUTION]: The instructions guide the agent to manage recurring tasks by creating or deleting scheduled jobs (e.g., cron jobs) based on observed activity. While this implies the use of system-level commands, the instructions describe the logic rather than providing hardcoded or dangerous command strings.
- [INDIRECT_PROMPT_INJECTION]: The skill defines an attack surface where the agent reads from a local file ('cron-tuner-state.json').
- Ingestion points: The agent reads state from 'cron-tuner-state.json' at the start of each run.
- Boundary markers: No specific delimiters or safety instructions are defined for the JSON data.
- Capability inventory: The skill implies file read/write access and the ability to modify system schedules.
- Sanitization: No explicit sanitization of the JSON content is mentioned, though the data is expected to be a simple schema controlled by the agent itself.
Audit Metadata