The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses a local Python script scripts/build_article_inventory.py to index files within the user-specified source directory. Analysis of the script confirms it only uses Python standard libraries (argparse, csv, json, re, pathlib) and performs local read/write operations without any network activity or privilege escalation.
[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and analyze untrusted content assets provided by the user (Markdown, Word, PDF, etc.).
Ingestion points: Document files found in the user-provided <源目录> as specified in SKILL.md.
Boundary markers: The instructions do not define strict delimiters (e.g., XML tags or special markers) to separate the untrusted document content from the agent's system instructions during processing.
Capability inventory: The skill possesses the ability to read arbitrary files in the target directory, execute a local Python script, and write output files to a 书稿策划输出/ subdirectory.
Sanitization: The build_article_inventory.py script implements basic sanitization by stripping Markdown syntax (code blocks, images, links) when generating text previews for its inventory files.
[SAFE]: No hardcoded credentials, malicious remote code downloads, or obfuscation techniques were detected in the skill files or scripts.

article2book