The Agent Skills Directory

[COMMAND_EXECUTION]: The skill coordinates its operations through several local Python scripts (food_fetch_setup.py, fetch_oa.py, resolve_oa.py, privacy_scan.py) to manage user configuration, resolve article status, and scan for privacy issues in deliverables.
[EXTERNAL_DOWNLOADS]: The skill downloads academic papers from well-known and trusted academic services including NCBI E-utilities, Unpaywall, OpenAlex, and arXiv. These are documented as official sources for legal open-access content.
[DATA_EXFILTRATION]: The skill matches requested references against the user's local reference manager folders (EndNote, Zotero, Mendeley). It transmits metadata (DOIs and titles) to external resolution APIs to determine availability.
[PROMPT_INJECTION]: This skill possesses an indirect prompt injection surface as it ingests untrusted data from academic articles (PDF, XML, HTML). It implements a content_reader agent to normalize this content into structured Markdown and explicitly forbids the bypass of safety or authentication mechanisms in its boundaries.md policy.

food-fetch