The Agent Skills Directory

[SAFE]: The skill consists of instructional markdown files and local Python scripts. A thorough review of the instructions and code confirms that the skill performs legitimate statistical operations without any malicious patterns, such as prompt injection, credential exfiltration, or unauthorized command execution. The referenced dependencies are well-established libraries in the data science community.- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided data structures for analysis, which creates a potential surface for indirect injection. 1. Ingestion points: The data parameter in functions within scripts/assumption_checks.py is used to intake external dataframes. 2. Boundary markers: The skill does not provide explicit boundary markers or instructions to ignore potential commands embedded in data. 3. Capability inventory: The skill is limited to statistical computation and data visualization using scipy, statsmodels, and pingouin, with no network access or file-writing capabilities. 4. Sanitization: No specific sanitization of the input data is performed. The risk is assessed as safe given the non-exploitable nature of the skill's capabilities.

statistical-analysis