auditing-cis-benchmark

Pass

Audited by Gen Agent Trust Hub on Jul 22, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSCREDENTIALS_UNSAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes various system commands such as systemctl, ps, ls, grep, and ulimit to audit operating system configurations, process isolation, and service statuses as required by the CIS Benchmark.
  • [EXTERNAL_DOWNLOADS]: It fetches binary checksums from binaries.cockroachdb.com to verify the integrity of the installed software. This is a trusted vendor domain and the operation is part of the standard auditing process.
  • [CREDENTIALS_UNSAFE]: The skill includes procedures to inspect certificate permissions and verify password hashing algorithms. One specific audit procedure utilizes the allow_unsafe_internals cluster setting to inspect stored password hashes to ensure they use SCRAM-SHA-256; this is documented as a restricted, read-only audit-only operation.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 22, 2026, 03:58 PM
Security Audit — agent-trust-hub — auditing-cis-benchmark