The Agent Skills Directory

[SAFE]: The skill follows secure coding practices for managing sensitive information. It instructs users to load the NCBI_API_KEY from environment variables rather than hardcoding them in scripts, which aligns with standard security best practices.
[COMMAND_EXECUTION]: The skill provides templates for executing external bioinformatics tools (e.g., BLAST+, MUSCLE, Clustal Omega) using the subprocess module. It includes explicit security warnings against interpolating unsanitized user input into shell commands and recommends using fixed argument lists to mitigate command injection risks.
[DATA_EXPOSURE]: The skill facilitates automated access to NCBI databases for sequence and publication data. These network operations are directed at well-known, trusted domains (ncbi.nlm.nih.gov) required for the skill's primary functionality.
[INDIRECT_PROMPT_INJECTION]: As the skill is designed to parse external biological data formats (FASTA, GenBank, XML), there is a theoretical surface for indirect injection from untrusted source data. This risk is managed through the use of mature, structured parsers provided by the Biopython library and instructions to validate data provenance.

biopython