genomic-intelligence
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFECREDENTIALS_UNSAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [CREDENTIALS_UNSAFE]: The skill instructions emphasize secure handling of API keys using environment variables (GI_API_KEY) and explicitly warn against hardcoding secrets in source code.\n- [EXTERNAL_DOWNLOADS]: The skill facilitates the fetching of genomic data from Ensembl's public REST API (rest.ensembl.org), which is a reputable and standard source in the genomic research field.\n- [INDIRECT_PROMPT_INJECTION]:\n
- Ingestion points: DNA sequences and genomic regions are ingested from user inputs and external databases (Ensembl).\n
- Boundary markers: None explicitly defined; however, the primary data format (DNA sequences) consists of nucleotide characters (ACGT) which have a very low risk of containing hidden instructions.\n
- Capability inventory: The skill utilizes the requests library for outbound network calls to Genomic Intelligence's inference API.\n
- Sanitization: DNA sequences are stripped of whitespace but no natural-language sanitization is applied, which is appropriate for genomic data types.
Audit Metadata