The Agent Skills Directory

[SAFE]: The skill interacts with the National Center for Biotechnology Information (NCBI) using official API endpoints (eutils.ncbi.nlm.nih.gov and api.ncbi.nlm.nih.gov). These are well-known, trusted scientific resources.
[SAFE]: The Python scripts (query_gene.py, fetch_gene_data.py, batch_gene_lookup.py) use standard libraries such as urllib.request and implement appropriate rate-limiting delays to comply with API usage policies.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from an external source (NCBI), which could theoretically contain instructions intended to influence the agent. However, given the source is a curated scientific database, this represents a standard informational surface with no specific risk detected.
Ingestion points: Data enters the agent's context through the output of scripts that query NCBI Gene metadata and descriptions.
Boundary markers: None; script outputs are provided as structured JSON or human-readable text without explicit delimiters.
Capability inventory: The skill is limited to making network requests to NCBI and reading local files; it has no capabilities for shell execution, local file modification, or credential exfiltration.
Sanitization: Search queries are properly handled using urllib.parse.urlencode to ensure safe transmission of parameters to the remote API.

gene-database