The Agent Skills Directory

[COMMAND_EXECUTION]: The skill involves the execution of a local Python script scripts/tracking_commit.py to handle project state synchronization and render Markdown views. Analysis of the script confirms it uses only standard Python libraries for local file operations and data validation without performing network requests or executing arbitrary external code.- [INDIRECT_PROMPT_INJECTION]: The skill handles user-provided text as part of the novel import process.\n- Ingestion points: Raw text content provided by the user in SKILL.md (Stage 1).\n- Boundary markers: The workflow uses rigid structural mappings to parse text into specific directories (正文/, 设定/, 大纲/).\n- Capability inventory: Local file writing and state management via the included Python script.\n- Sanitization: The Python script enforces strict schema validation and byte-size limits on processed data.- [DATA_EXPOSURE]: The skill accesses project-specific configuration files such as .story-deployed. These files are internal to the moshu project structure and do not contain sensitive system credentials or private keys.

moshu-import