gget
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted biological data files and sequence strings which could contain adversarial instructions targeting the agent. 1. Ingestion points: Modules such as gget blast, gget muscle, and gget mutate ingest external data from FASTA, CSV, and TSV files, as well as direct sequence strings. 2. Boundary markers: There are no instructions for the agent to use delimiters or ignore instructions embedded within the processed biological data. 3. Capability inventory: The skill includes the ability to write files to the local file system using the -o/--out flag and performs network operations to genomic databases and the OpenAI API. 4. Sanitization: The skill does not mention any validation or sanitization mechanisms for the content of the data files before processing.
- [EXTERNAL_DOWNLOADS]: The skill includes setup commands that fetch significant amounts of data and dependencies from external sources. Evidence: The gget setup module downloads approximately 4GB of model parameters for AlphaFold and local databases for the Eukaryotic Linear Motif (ELM) analysis. These downloads are functional requirements for the tool's intended scientific use cases.
Audit Metadata