hmdb-database
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to parse a massive external XML file (hmdb_metabolites.xml), which may contain untrusted text content that is then processed by the AI agent. This establishes an attack surface for indirect prompt injection if the source data contains malicious instructions.
- Ingestion points: The skill utilizes Python's xml.etree.ElementTree to parse the local 'hmdb_metabolites.xml' file, loading its content into the agent's context.
- Boundary markers: The provided implementation and instructions do not include specific delimiters or 'ignore embedded instructions' warnings for the data extracted from the XML.
- Capability inventory: The skill possesses file system read access for the XML file and write access to generate CSV exports, alongside standard data manipulation using pandas.
- Sanitization: There is no evidence of text sanitization, validation, or escaping of the metabolite descriptions or other fields extracted from the XML before they are processed by the agent.
Audit Metadata