The Agent Skills Directory

[SAFE]: The skill uses localized Python and Node.js scripts for project maintenance and validation. These scripts (e.g., tracking_commit.py, check-ai-patterns.js) only interact with files within the user's project directory and do not perform any network operations or access sensitive system data.\n- [SAFE]: Data ingestion is limited to benchmark novels and project settings stored in the workspace. While processing external text files is a potential surface for indirect prompt injection, the skill's design does not include capabilities that would allow such content to compromise the host system or the agent's core instructions.\n- [SAFE]: All external links and metadata point to the author's official GitHub repository, and no instances of obfuscation or dynamic code execution from untrusted sources were detected.

story-long-write