cron-schedule

Warn

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions and examples for using the CronCreate tool to establish workers that survive session restarts. This functionality allows for the creation of persistent background processes that operate independently of the active user session, which could be used to maintain unauthorized access or execute hidden tasks.
  • [PROMPT_INJECTION]: The skill utilizes a pattern where natural language prompts are scheduled for delayed execution via CronCreate. This creates a surface for indirect prompt injection where scheduled instructions could be manipulated or lack necessary context.
  • Ingestion points: The prompt parameter within the CronCreate tool call in SKILL.md.
  • Boundary markers: Absent; there are no delimiters or instructions to ignore embedded commands within the scheduled strings.
  • Capability inventory: Includes CronCreate, CronList, CronDelete, and mcp__claude-flow__hooks_worker-dispatch (SKILL.md).
  • Sanitization: Absent; the skill does not define validation or filtering for the content of scheduled prompts.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 13, 2026, 01:22 PM
Security Audit — agent-trust-hub — cron-schedule