The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes the local Python script scripts/eda_analyzer.py and suggests generating custom Python analysis code to process user-provided scientific data.
[EXTERNAL_DOWNLOADS]: The skill refers users to install standard scientific Python libraries (such as biopython, rdkit, and nd2reader) and provides a link to the author's research platform, K-Dense Web (k-dense.ai).
[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it is designed to parse and summarize content from untrusted external scientific data files.
Ingestion points: The agent reads data from user-provided files in various formats (e.g., .fastq, .pdb, .csv, .nd2).
Boundary markers: There are no specific delimiters or instructions defined in the workflow to prevent the agent from following instructions embedded within the data files.
Capability inventory: The skill uses Python scripts and specialized scientific libraries to extract metadata and perform statistical analysis on the ingested content.
Sanitization: The skill instructions do not include steps for sanitizing or validating the data from the files before it is processed by the agent.

exploratory-data-analysis