claude-code
Pass
Audited by Gen Agent Trust Hub on Sep 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill makes extensive use of the
terminal()tool to executeclaudeCLI commands andtmuxfor session management. These are documented as the primary intended function of the skill for task delegation. - [EXTERNAL_DOWNLOADS]: The skill provides instructions for installing the official
@anthropic-ai/claude-codepackage from npm and usingclaude update/upgradeto maintain the CLI tool. These are directed at official Anthropic services. - [SAFE]: Static detector flags for fork bombs were evaluated as false positives; the content includes a shell snippet example of a dangerous command string inside a security hook description designed to block such attacks, rather than executing them. The use of
--dangerously-skip-permissionsis clearly documented as a safety bypass requiring explicit user orchestration.
Audit Metadata