similar-protein-retrieval
Warn
Audited by Snyk on Mar 25, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.70). This skill fetches and ingests content at runtime from public third-party databases (e.g., query_uniprot calls https://rest.uniprot.org/uniprotkb/{id}, query_pdb_async downloads PDB files from https://files.rcsb.org, and it invokes FoldSeek/MSA requesters against UniRef/AFDB), and that remote content is parsed and used to drive search choices and downstream actions, exposing the agent to untrusted external data that could influence behavior.
Issues (1)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata