The Agent Skills Directory

[Prompt Injection] (SAFE): The skill contains only functional instructions for time series decomposition and lacks any patterns suggesting intent to bypass safety filters or override system prompts.
[Data Exposure & Exfiltration] (SAFE): No hardcoded credentials, sensitive file paths, or suspicious network operations were identified. Data access is restricted to user-provided CSV files.
[Unverifiable Dependencies] (SAFE): All listed dependencies (pandas, numpy, scipy, statsmodels, matplotlib) are well-known, reputable packages from the official Python Package Index (PyPI).
[Indirect Prompt Injection] (LOW): The skill processes external data via CSV files. While this is an ingestion surface, the processing is handled by structured data libraries for mathematical analysis, presenting a negligible risk of malicious data influencing agent behavior.
Ingestion points: load_csv method in ts_decomposer.py (referenced in SKILL.md).
Boundary markers: None explicitly defined in the API description.
Capability inventory: File writing via plot_components and plot_seasonal.
Sanitization: Relies on standard pandas CSV parsing which treats inputs as data values rather than instructions.

time-series-decomposer