etetoolkit

Warn

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS] (MEDIUM): The skill scripts depend on the 'ete3' library, which is an external Python package not listed among trusted sources. This introduces a dependency on unverified third-party code.
  • [PROMPT_INJECTION] (MEDIUM): The skill is susceptible to indirect prompt injection via the processing of untrusted Newick tree files and taxa lists. Ingestion points: 'args.input' in both scripts and 'args.keep_taxa' in 'tree_operations.py'. Boundary markers: Absent. Capability inventory: 'tree.render' in 'quick_visualize.py', 'tree.write' in 'tree_operations.py', and console output. Sanitization: Minimal, limited to basic output file extension validation.
  • [COMMAND_EXECUTION] (LOW): The provided scripts are intended for command-line execution. While consistent with the stated purpose, use of such tools by an agent on untrusted inputs requires careful monitoring.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 16, 2026, 01:01 PM