Skills
Navigate
skills/khendzel/skills-janitor/janitor-value/Gen Agent Trust Hub

janitor-value

Pass

Audited by Gen Agent Trust Hub on Jul 21, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute local scripts (value.sh and mcp.sh) intended to analyze skill performance and token usage. These scripts are invoked from the local filesystem within the agent's configuration directory.
  • [SAFE]: The skill reads local metadata and session history from ~/.claude/skills and ~/.claude/agents. This data access is strictly limited to the local environment for reporting purposes, with no evidence of exfiltration or network communication detected.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 21, 2026, 05:53 PM
Security Audit — agent-trust-hub — janitor-value