biopython

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The prompt shows and encourages hardcoding Entrez.api_key (and an example email) into code, which instructs embedding API credentials verbatim in generated output and therefore risks exposing secrets.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.80). The skill's workflow explicitly instructs the agent to fetch and parse open, user-populated third‑party resources (e.g., NCBI via Bio.Entrez in references/databases.md and references/blast.md using Entrez.efetch and NCBIWWW.qblast, and RCSB PDB via PDBList.retrieve_pdb_file in references/structure.md), so untrusted external content is ingested and used to drive subsequent parsing and actions.