ai-drama-prompt-factory
AdvisoryAudited by Static analysis on Apr 30, 2026.
Overview
No suspicious patterns detected.
Findings (0)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
If pointed at a broad folder such as a home directory or workspace, the skill could store and later reuse snippets or keywords from private documents that were not intended for the short-drama project.
The script can create a persistent index for an arbitrary file or directory and stores previews/keywords derived from file contents. The artifacts do not clearly bound which paths may be indexed or how the index should be retained or deleted.
python3 doc_indexer.py <文档路径或目录> [--output index.json] [--update] ... current_section['content_preview'] ... 'keywords': extract_keywords(content)
Only index a dedicated project/source folder containing the story text. Avoid indexing broad personal or work directories, review the generated index before reuse, and delete it when the project is done.
The agent may read substantial portions of indexed files; if the index contains unintended files, their contents could enter the working context or outputs.
The fetch helper can retrieve content from any file recorded in the index. This is expected for long-document adaptation, but it depends on the index being limited to intended source material.
python3 doc_fetch.py --index doc_index.json --file "文件名" --all ... def fetch_all(filepath: str, max_lines: int = 500)
Confirm the target file or section before fetching, and keep the index limited to the novel/story files needed for the project.