ensembl-database
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWEXTERNAL_DOWNLOADS
Full Analysis
- Unverifiable Dependencies & Remote Code Execution (LOW): The skill utilizes the
ensembl_restandrequestsPython packages. These are standard libraries for bioinformatics and HTTP operations; their use is appropriate for the skill's stated purpose. - Data Exposure & Exfiltration (INFO): The skill communicates with official Ensembl REST API endpoints (e.g.,
rest.ensembl.org). No sensitive credentials or private user files are accessed or transmitted. - Indirect Prompt Injection (LOW): The skill ingests data from external scientific databases.
- Ingestion points: Data enters the system via the
requests.getandensembl_restclient calls to external API endpoints. - Boundary markers: None are explicitly used in the provided logic.
- Capability inventory: The skill only performs network read operations. It lacks file-system write access, subprocess execution, or administrative capabilities.
- Sanitization: No specific sanitization of the genomic data is performed, but given the lack of downstream execution capabilities, the risk is minimal.
- Prompt Injection (SAFE): There are no instructions in the metadata or body attempting to override agent behavior, extract system prompts, or bypass safety filters.
Audit Metadata