The Agent Skills Directory

[SAFE]: The skill is designed for data analysis using well-known, trusted Python libraries (pandas, scipy, statsmodels, numpy). No malicious patterns such as credential theft, unauthorized network access, or privilege escalation were detected.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external datasets (CSV, Excel, Parquet, JSON). While this creates an ingestion surface, the risk is minimal as the instructions focus on quantitative statistical methods rather than the interpretation of natural language instructions within the data.
Ingestion points: Dataset file paths provided to the agent and loaded via pd.read_csv (seen in SKILL.md).
Boundary markers: Absent.
Capability inventory: Python data manipulation and statistical execution.
Sanitization: Absent.

data-analysis