The Agent Skills Directory

[SAFE]: The skill provides legitimate bioinformatics processing patterns using the standard biopython library.
[COMMAND_EXECUTION]: Uses multiprocessing and concurrent.futures for CPU-intensive data processing. These are implemented correctly using static function references and do not involve shell injection or execution of untrusted commands.
[DATA_EXPOSURE]: All file operations are directed toward local data directories (e.g., data/, input/, output/). There is no evidence of network exfiltration or access to sensitive configuration files like .ssh or .aws credentials.
[INDIRECT_PROMPT_INJECTION]: The skill processes external biological sequence data. While sequence headers (e.g., FASTA descriptions) are untrusted inputs, the skill treats them as data for labeling or organization and does not interpolate them into prompts as instructions, maintaining a safe boundary between data and logic.

bio-batch-processing