The Agent Skills Directory

[SAFE]: Detailed analysis of the skill's instructions and scripts reveals no malicious patterns or security risks. The skill is dedicated to retrieving public biological data from a reputable source.
[EXTERNAL_DOWNLOADS]: The skill makes HTTP requests exclusively to rest.uniprot.org. This is the official API endpoint for UniProt, a well-known and trusted international resource for protein sequence and functional information.
[COMMAND_EXECUTION]: The scripts/uniprot_client.py script is a standard Python utility. It uses the requests library for network communication and argparse for its command-line interface. It does not perform any dangerous system calls, shell command execution, or file system modifications beyond writing data to a specified output file when requested by the user.
[DATA_EXFILTRATION]: The skill does not access sensitive local files (such as .ssh, .aws, or .env), credentials, or environment variables. It only transmits search queries and identifiers to the UniProt API.
[PROMPT_INJECTION]: The instructions in SKILL.md are purely technical and do not contain any patterns intended to override agent safety guidelines or manipulate underlying model behavior.

uniprot-database