tool-hooks-doctor
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEPERSISTENCEDYNAMIC_EXECUTIONCOMMAND_EXECUTION
Full Analysis
- [PERSISTENCE]: The skill modifies
~/.claude/settings.jsonand.claude/settings.jsonto injecthooks. These hooks are configured to automatically execute bash scripts (such aspre-tool.shandpost-bash.sh) located in the user's home directory whenever specific tools are used or a session ends. This creates a persistence mechanism where code is executed across different AI agent sessions. - [DYNAMIC_EXECUTION]: The skill generates and executes a Python script via a bash heredoc (`python3
- <<'PY'`) to perform complex JSON manipulation on settings files. This allows the skill to programmatically update configuration files rather than just appending text, but it involves runtime script generation and execution.
- [COMMAND_EXECUTION]: The skill uses
bashto perform system diagnostics, including listing directory contents (ls -la) and searching through sensitive configuration files (grep) to identify the current status of the evolution hooks. These commands are used for read-only verification of the environment.
Audit Metadata