The Agent Skills Directory

[SAFE]: The skill implements a complex writing coach workflow. It relies on a well-defined project structure and several helper scripts to maintain consistency and quality.
[COMMAND_EXECUTION]: The Python script storyctl.py uses subprocess.run to invoke local Node.js linting scripts (check-ai-patterns.js, check-degeneration.js, etc.). These operations are used to verify the output of the writing process and do not involve shell injection or execution of arbitrary code.
[DYNAMIC_EXECUTION]: Local Python modules are loaded via importlib in storyctl.py and tracking_commit.py. This is used to share core logic between scripts and is restricted to files within the skill's own directory.
[DATA_EXPOSURE]: The skill manages an 'author memory' and story state within the project's workspace (e.g., in .story/作者记忆/ and 追踪/). These operations are consistent with the skill's purpose of tracking a long-form narrative's continuity and the author's stylistic preferences.

story-long-write