skills/jakkaj/tools/crew-cut/Gen Agent Trust Hub

crew-cut

Pass

Audited by Gen Agent Trust Hub on Jun 25, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists entirely of natural language instructions for the agent to act as a reviewer or task-primer. It does not include any executable scripts, binary files, or configuration that would allow for malicious actions.
  • [DATA_EXFILTRATION]: No network tools or data exfiltration patterns were identified. The instructions state the skill is 'read-only' and forbids routing or continuing the workflow beyond returning the review/summary.
  • [COMMAND_EXECUTION]: There are no shell commands, subprocess calls, or privilege escalation attempts. The skill specifically instructs the agent not to execute or edit code.
  • [INDIRECT_PROMPT_INJECTION]: While the skill is designed to process external 'targets' (code, docs, etc.), it lacks the capabilities (such as network access or file writing) to exploit any malicious instructions contained within that data. It also explicitly instructs the agent to preserve 'safety' and 'correctness'.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 25, 2026, 01:41 AM
Security Audit — agent-trust-hub — crew-cut