The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external marketing data from sources such as CSV files to build statistical models.
Ingestion points: Media and target data are loaded via pandas.read_csv in SKILL.md and references/data_analysis.md.
Boundary markers: The instructions do not define specific delimiters or instructions for the agent to ignore potentially malicious content within the processed data.
Capability inventory: The skill includes capabilities for file operations (mmm.save), data visualization (plotting), and fitting complex Bayesian models.
Sanitization: The skill relies on standard data validation, preprocessing, and scaling routines provided by the pymc-marketing framework.
[DYNAMIC_EXECUTION]: The custom model implementation uses the patsy library for generating spline bases, which involves the evaluation of string-based formulas.
Evidence: The skill demonstrates the use of patsy.dmatrix with formula strings in references/custom_model.md.
Context: Formula evaluation is a standard component of statistical modeling in Python and is utilized here for its intended purpose without unsafe interpolation of external input.

mmm-modeling