The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/enrichgo_runner.py utilizes subprocess.run to execute an R script dynamically generated from a template. This execution is necessary to perform Gene Ontology analysis using the clusterProfiler R package, which is a standard requirement for the skill's stated scientific purpose.
[EXTERNAL_DOWNLOADS]: The skill documentation describes the installation of R package dependencies and relies on the gseapy Python library, which fetches genomic pathway data from the Enrichr API. These network operations are functional requirements for pathway enrichment analysis.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect injection as it processes untrusted user-supplied data. Ingestion points: Data enters through gene list files, background universe files, and pre-existing scientific notebooks. Boundary markers: The instructions emphasize using pre-computed results and specific thresholds to minimize interpretation variance but do not define explicit delimiters for raw gene list data. Capability inventory: The skill possesses the ability to execute shell commands (via Rscript) and write to the local file system (specifically the /tmp directory). Sanitization: The scripts perform standard data cleanup, such as stripping headers, normalizing whitespace, and removing version suffixes from Ensembl IDs.

tooluniverse-gene-enrichment