The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/content_extract.py executes a local wrapper script (mineru_parse_documents.py) using subprocess.run. The command is constructed as a list, which avoids shell injection risks.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from arbitrary external URLs provided by the user. (1) Ingestion points: The url parameter in SKILL.md and scripts/content_extract.py triggers network fetching via web_fetch or a local tool. (2) Boundary markers: The instructions do not define specific delimiters to isolate the fetched content, although the output is returned in a structured JSON format. (3) Capability inventory: The skill has network access and can execute a local extraction script via subprocess. (4) Sanitization: The extracted content is capped at 20,000 characters and returned within a JSON structure.

content-extract