alterlab-etetoolkit
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill facilitates the download of a large taxonomy database (~300MB) from the National Center for Biotechnology Information (NCBI) via the
ete3.NCBITaxaclass. This is a standard procedure for local taxonomy lookup and utilizes a well-known scientific service domain. - [COMMAND_EXECUTION]: The skill includes Python scripts (
scripts/tree_operations.pyandscripts/quick_visualize.py) intended for command-line use. These scripts allow the agent to execute complex tree manipulation and rendering tasks using provided file paths and parameters. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and parse external data files in Newick and NHX formats, creating an indirect prompt injection surface.
- Ingestion points: External files are loaded into the agent's context through the
TreeandPhyloTreeconstructors in the provided scripts and documentation examples. - Boundary markers: The skill does not provide explicit delimiters or instructions to the agent to ignore potentially malicious content embedded within the scientific metadata of the tree files.
- Capability inventory: The skill has capabilities for file system writes (
tree.write,tree.render) and shell execution via provided utility scripts. - Sanitization: The skill relies on standard Newick parsing logic and does not implement additional validation or filtering of string content within the trees before it is processed by the agent.
Audit Metadata