biopython-advanced
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes various external biological data formats such as FASTA, GenBank, and GenePop, which introduces a surface for indirect prompt injection attacks.\n
- Ingestion points: External data is ingested via SeqIO.read, GenePop.read, and motifs.read in the documented scripts.\n
- Boundary markers: No explicit instruction delimiters or boundary markers are employed to isolate data from potential instructions.\n
- Capability inventory: Capabilities are restricted to bioinformatics calculations and writing results to local JSON and PDF files; no shell execution or network exfiltration patterns were found.\n
- Sanitization: Input data is handled by standard library parsers without specific filtering for natural language prompts.\n- [SAFE]: Fetches biological data from NCBI's PubChem database using the Entrez API, which is a well-known scientific service.\n- [SAFE]: Uses the deprecated Bio.pairwise2 module for sequence alignment, which is a best-practice violation rather than a security risk.
Audit Metadata