The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill provides instructions and code templates that utilize the pickle module for object serialization and model persistence.
Evidence: Found in references/workflows.md and referenced in SKILL.md, the code demonstrates saving and loading model objects using pickle.dump() and pickle.load().
Risk: The pickle module is known to be insecure against maliciously constructed data. Executing pickle.load() on an untrusted file can lead to arbitrary code execution within the agent's environment.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to the way it processes external data and its available capabilities.
Ingestion points: The skill templates in assets/hierarchical_model_template.py and assets/linear_regression_template.py ingest untrusted data from local CSV files via pd.read_csv().
Boundary markers: There are no explicit boundary markers or instructions to the model to ignore embedded instructions within the ingested datasets.
Capability inventory: The skill possesses extensive file system write capabilities, including generating diagnostic plots (plt.savefig), exporting summary statistics (summary.to_csv), and saving binary model results (idata.to_netcdf).
Sanitization: No data validation, escaping, or sanitization logic is implemented for external content before it is processed by the analysis scripts.

pymc