The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests external data, which introduces a potential attack surface for indirect prompt injection.
Ingestion points: The file examples/go_all_ontologies.R reads gene information from an external de_results.csv file.
Boundary markers: There are no explicit delimiters or warnings provided to the agent to ignore instructions potentially embedded within the input data.
Capability inventory: The skill's capabilities are limited to local bioinformatics computations and writing output files to a temporary directory (tempdir()). It does not have access to the network, sensitive file paths, or the ability to execute arbitrary shell commands.
Sanitization: No explicit validation or sanitization is performed on the CSV contents before they are processed by R functions.
[SAFE]: The skill uses established and trusted bioinformatics libraries (clusterProfiler, org.Hs.eg.db, goseq) from the well-known Bioconductor repository.
[SAFE]: All file operations are restricted to local data processing and temporary file storage, with no evidence of data exfiltration or unauthorized system access.
[SAFE]: No obfuscation techniques, hidden URLs, or malicious persistence mechanisms were detected in any of the provided files.

bio-pathway-go-enrichment