skills/boshu2/agentops/flywheel/Gen Agent Trust Hub

flywheel

Pass

Audited by Gen Agent Trust Hub on Apr 5, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses local shell commands (find, ls, grep, awk, jq) and a maintenance utility called ao to calculate knowledge metrics and perform automated cleanup tasks like deduplication and archiving. All operations are restricted to the local environment.
  • [SAFE]: Filesystem access is scoped to the .agents/ workspace and the ~/.claude/patterns/ directory. This access is consistent with the skill's purpose of monitoring knowledge flywheel health and is not used for sensitive data harvesting.
  • [SAFE]: The skill includes a reference consistency checker that parses markdown files to identify broken internal links. The parsing logic is implemented using a safe regular expression within an awk script.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 5, 2026, 11:06 PM