The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill facilitates data retrieval from established and trusted scientific repositories, including NCBI Taxonomy, UniProt, ChEMBL, and various literature databases like PubMed and BioRxiv. These references are documented neutrally as they represent standard scientific services.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface due to the processing of untrusted external literature data. Ingestion points: Data is collected from external sources via tools such as UniProt_search, PubMed_search_articles, and BioRxiv_search_preprints. Boundary markers: The skill does not implement specific boundary markers or delimiters for the ingested text in its generated research reports. Capability inventory: The skill is capable of writing research findings to local markdown and CSV files and performing further tool calls based on initial research results. Sanitization: There is no evidence of explicit sanitization or filtering of content retrieved from external databases before it is integrated into the output.
[SAFE]: No high-risk security patterns, such as hardcoded credentials or malicious command execution, were identified. The skill's architecture is consistent with its intended purpose of biomedical research, and it correctly utilizes environment variables for API configuration.

tooluniverse-infectious-disease