skills/goldengrape/agent-company/cron/Gen Agent Trust Hub

cron

Pass

Audited by Gen Agent Trust Hub on Feb 20, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • Prompt Injection (LOW): The skill is susceptible to indirect prompt injection via the 'Task' mode, which instructs the agent to execute the 'message' parameter as a task description. This could allow an attacker to inject recurring malicious tasks if the agent processes data from external sources. (1) Ingestion points: The 'message' parameter in the cron(action='add') tool. (2) Boundary markers: None are provided to separate user instructions from untrusted data. (3) Capability inventory: The agent is explicitly told to execute the message and report the result, which can include network or file system operations. (4) Sanitization: No sanitization or validation of the task content is defined.
  • Persistence (LOW): As a tool designed for recurring tasks, the skill inherently provides a mechanism for persistence. While this is the intended primary purpose, it could be leveraged to maintain malicious logic within the agent's environment across sessions.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 20, 2026, 05:25 PM