skills/pivanov/ai-skills/split-tasks/Gen Agent Trust Hub

split-tasks

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • Prompt Injection (SAFE): No malicious instructions, bypass attempts, or system prompt extraction techniques were found. The prompt logic is focused on task decomposition and operational efficiency.
  • Data Exposure & Exfiltration (SAFE): No hardcoded credentials, sensitive file paths (e.g., SSH keys, AWS configs), or unauthorized network operations are present in the skill definition.
  • Unverifiable Dependencies & RCE (SAFE): The skill does not include package installations, remote script downloads, or the execution of external code. It orchestrates existing agent types without adding new execution vectors.
  • Indirect Prompt Injection (LOW): The skill is designed to process untrusted user input and delegate it to sub-agents with significant capabilities (like Bash for command execution). While this creates an attack surface where a user could provide a task that exploits a sub-agent, the skill itself provides safety 'Red Flags' to prevent file-system conflicts and does not contain malicious logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:29 PM