The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill fetches paper metadata and citation rankings from well-known scholarly APIs, including DBLP (dblp.org) and Semantic Scholar (api.semanticscholar.org).
[COMMAND_EXECUTION]: The agent executes included Python scripts (lookup_exemplar.py, rank_top_cited.py, and build_exemplar_bundle.py) to automate data collection and analysis synthesis.
[DATA_EXFILTRATION]: The CONTACT_EMAIL environment variable is transmitted in the User-Agent header of HTTP requests to identify the user to scholarly API providers, which is a standard requirement for 'polite' access to these services.
[PROMPT_INJECTION]: The skill processes untrusted external content from research papers; however, the indirect prompt injection risk is significantly mitigated by the skill's design, which emphasizes transient processing and focuses on structural metadata rather than executing instructions found within the text.

study-exemplars