NYC

biopython

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [Metadata Analysis] (SAFE): The ID, name, author, and description fields were inspected and found to be free of prompt injections, malicious instructions, or obfuscation.
  • [Indirect Prompt Injection Surface] (LOW): The skill is designed to process external biological data and query NCBI/PubMed APIs. This constitutes an indirect prompt injection surface as untrusted data from these sources could contain instructions meant to influence the LLM. Ingestion points: NCBI queries, FASTA/GenBank file parsing. Boundary markers: Unknown (code missing). Capability inventory: Network access (Bio.Entrez), file read/parse. Sanitization: Unknown.
  • [Analysis Limitation] (SAFE): Only the metadata.json file was provided. The 'SKILL.md' file and any associated scripts were missing, limiting the scope of this security audit to metadata only.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:28 PM