cron
Pass
Audited by Gen Agent Trust Hub on Apr 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill defines a 'Task' mode that allows the agent to execute instructions stored in the 'message' parameter. This enables recurring execution of agent tasks on a defined schedule.
- [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection because the agent executes the contents of the 'message' parameter. If this parameter is populated with untrusted data from external sources, it could lead to the persistence of unauthorized tasks.
- Ingestion points: The 'message' parameter in the 'cron' tool call in SKILL.md.
- Boundary markers: None identified in the skill definition.
- Capability inventory: Scheduled, recurring execution of agent instructions (tasks) with result reporting.
- Sanitization: No explicit sanitization or validation of the 'message' content is described in the tool definition.
Audit Metadata