The Agent Skills Directory

[Data Exposure & Exfiltration] (LOW): The skill performs network requests to rest.uniprot.org. While this is the intended functionality for biological data retrieval, the domain is not on the trusted whitelist for network operations.
[Indirect Prompt Injection] (LOW): The skill ingests untrusted data from the UniProt REST API, which is a potential surface for indirect injection if the source data were compromised.
Ingestion points: API response handling in scripts/uniprot_client.py (functions search_proteins, get_protein, map_ids).
Boundary markers: Absent; external data is returned directly to the agent context.
Capability inventory: Network read access to UniProt; stdout printing. No write operations or command execution capabilities are present.
Sanitization: The skill uses the requests library, which safely handles URL parameter encoding and JSON parsing.
[Unverifiable Dependencies] (LOW): The uniprot_client.py script relies on the requests Python package. No requirements.txt or version pinning is provided to ensure supply chain integrity.

uniprot-database