skills/rawknee-69/beta-claw/cron/Gen Agent Trust Hub

cron

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEPERSISTENCEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PERSISTENCE]: The skill provides instructions and examples for modifying the system crontab using the (crontab -l 2>/dev/null; echo "*/5 * * * * /path/to/script.sh") | crontab - pattern. This allows scripts to be executed at recurring intervals, maintaining execution across system reboots and sessions.
  • [INDIRECT_PROMPT_INJECTION]: The betaclaw schedule add command accepts an --instruction parameter (e.g., --instruction "Send a warm check-in message to the user"). This creates a surface where natural language instructions are stored and executed by the agent at a later time, potentially allowing for delayed injection if the source of the instruction is untrusted.
  • Ingestion points: Data enters the scheduler via the --instruction argument in SKILL.md examples.
  • Boundary markers: None identified in the provided command patterns.
  • Capability inventory: The skill uses the exec tool to perform shell operations and manage schedules.
  • Sanitization: No evidence of input sanitization or validation for the instruction string.
  • [COMMAND_EXECUTION]: The skill leverages the exec tool to run shell-level commands, specifically the crontab utility and the betaclaw CLI, to manage task scheduling and listing.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 12:30 PM
Security Audit — agent-trust-hub — cron