tooluniverse-sequence-analysis
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it is designed to ingest and process external biological data and analysis artifacts.
- Ingestion points: The agent is instructed to read pre-computed results from local files, executed Jupyter notebooks (*_executed.ipynb), and fetch sequence data from external databases including NCBI, Ensembl, and UniProt.
- Boundary markers: No specific delimiters or instructions to ignore embedded instructions are defined for the ingested data content.
- Capability inventory: The skill utilizes network operations via urllib in scripts/sequence_tools.py to fetch protein data and provides templates for shell-level execution of tools like samtools, BWA, and Trimmomatic.
- Sanitization: The skill does not explicitly include sanitization or validation logic for the content of retrieved sequences or local notebooks before they are processed by the agent.
- [EXTERNAL_DOWNLOADS]: The scripts/sequence_tools.py script includes functionality to fetch protein sequences from the official UniProt REST API (rest.uniprot.org). This is a well-known and trusted biological data service used for the skill's primary analysis purpose.
Audit Metadata