The Agent Skills Directory

[SAFE]: The skill uses established Harness MCP tools (harness_list, harness_describe) to access audit logs. These tools are restricted to read-only access for audit_event resources.
[SAFE]: No evidence of data exfiltration or hardcoded credentials was found. The skill processes sensitive audit data within the agent's context to generate user-requested reports.
[SAFE]: The skill does not perform any external downloads, remote code execution, or privilege escalation. It relies entirely on built-in tools and static templates.

audit-report