The Agent Skills Directory

[SAFE]: The skill performs legitimate bioinformatics tasks including genomic region annotation, gene ID conversion via the 'mygene' service, and GO/KEGG enrichment analysis.
[DATA_EXPOSURE]: The skill processes user-provided genomic data (BED files and gene lists) and creates local project directories for output. It uses the 'mygene' tool which connects to a well-known bioinformatics API (mygene.info) for ID mapping, which is consistent with its stated purpose and does not represent unauthorized data exfiltration.
[COMMAND_EXECUTION]: The skill orchestrates several tools (mcp__homer-tools, mcp__file-format-tools, mcp__project-init-tools) to execute shell-based bioinformatics pipelines. These operations are scoped to the project directory and are standard for the intended use case.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external files (BED and gene lists) and processes them using various tools. While it lacks explicit boundary markers or sanitization instructions for these inputs, the capabilities are restricted to bioinformatics processing and local file management, posing a low risk of exploitation within this context.

functional-enrichment