The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes Newick/NHX tree files and sequence alignments, which represents a surface for indirect prompt injection. Malicious instructions could be embedded in node names or sequence metadata within these external data files.\n
Ingestion points: Processes tree.nw and alignment.fasta using scripts like scripts/tree_operations.py and scripts/quick_visualize.py.\n
Boundary markers: The skill lacks explicit markers or instructions to the agent to ignore potentially malicious content within these data files.\n
Capability inventory: The skill allows for file-system writes (tree.write, tree.render) and network access for taxonomic database updates (NCBITaxa).\n
Sanitization: No data validation or sanitization is performed on input files beyond standard library parsing.\n- [EXTERNAL_DOWNLOADS]: The skill uses the NCBITaxa component to download and update a local taxonomic database (~300MB) from the National Center for Biotechnology Information (NCBI). This is a standard and documented feature for biological data integration from a well-known scientific service.\n- [COMMAND_EXECUTION]: Includes Python scripts (scripts/tree_operations.py, scripts/quick_visualize.py) that perform tree manipulation, format conversion, and rendering based on user-provided file paths and parameters.

etetoolkit