The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes several Python scripts (check_chapter_wordcount.py, generate_epub.py, check_novel_health.py, check_rhythm.py, and translate_to_english.py) to handle utility tasks such as word count verification, rhythm analysis, and e-book generation. These scripts operate on local files within the novel project directory.
[EXTERNAL_DOWNLOADS]: The translate_to_english.py script requires standard third-party libraries (openai, anthropic, and google-generativeai) to communicate with official AI service endpoints. These are well-known technology providers.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8) by reading and processing user-controlled Markdown files (drafts and outlines) to provide context for AI-driven writing and translation tasks.
Ingestion points: The agent reads project files located in the novels/ directory, including 00-大纲.md and chapter files.
Boundary markers: The skill relies on standard Markdown headers (e.g., ## 正文) to delimit content sections during processing.
Capability inventory: The skill executes Python scripts and performs file read/write operations within the project scope.
Sanitization: Novel content is ingested directly for processing without specialized escaping or sanitization filters.

chinese-novelist-skill