agent-ops

Pass

Audited by Gen Agent Trust Hub on May 5, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements standard SRE (Site Reliability Engineering) patterns for agent management, such as circuit breakers and cost caps.
  • [COMMAND_EXECUTION]: Uses standard shell commands like tmux, du, grep, and date for monitoring and managing agent sessions. These operations are limited to the local environment and the agent's own data directories.
  • [DATA_EXPOSURE]: Accesses metrics and configuration files within the agent's specific data directories (e.g., ~/.claude/metrics). No sensitive system files or credentials are accessed.
Audit Metadata
Risk Level
SAFE
Analyzed
May 5, 2026, 09:05 AM