tooluniverse-metabolomics
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill implements a research pipeline that ingests data from external metabolomics databases (such as HMDB and MetaboLights) and writes it directly into a markdown report. This creates a surface for indirect prompt injection if the external data were to contain malicious instructions.
- Ingestion points: Data is retrieved from external APIs via
ToolUniversetool calls inpython_implementation.pysuch asHMDB_search,metabolights_get_study, andMetabolomicsWorkbench_get_study. - Boundary markers: The generated reports do not utilize boundary markers or delimiters to isolate data retrieved from external sources.
- Capability inventory: The skill includes file-writing capabilities to save analysis reports to the local file system in
python_implementation.py. - Sanitization: There is no evidence of sanitization, escaping, or validation of the content retrieved from external databases before it is interpolated into the final report.
- [EXTERNAL_DOWNLOADS]: The skill fetches metabolite information, chemical properties, and study metadata from established scientific repositories including the Human Metabolome Database (HMDB), MetaboLights, and PubChem. These are well-known services provided by academic and governmental organizations.
Audit Metadata