The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to execute local Python scripts such as webnovel.py and reference_search.py. These scripts are part of the skill's own internal directory structure and are used for routine tasks like environment preflight, project path resolution, and reference searching.
[SAFE]: No signs of prompt injection, data exfiltration, or obfuscation were detected. The skill maintains a strict separation of concerns by delegating complex logic (like context generation and review) to sub-agents, which is a security best practice for autonomous systems.
[SAFE]: The workflow relies on local project data (outlines, settings, and state files) and does not perform any unauthorized external network operations. All environment variables used (CLAUDE_PROJECT_DIR, CLAUDE_PLUGIN_ROOT) are standard for the host platform.

webnovel-write