Menu
Run multi-step SQL plus Python/statistics/ML analysis while keeping tools thin and evidence grounded
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
On-demand time-series forecasting. CAPTURE params from project context, call run_forecast, present deterministic engine results.
Use when helping initialize, configure, or prepare a Seeknal project like a coding agent
Translate business questions into metrics, SQL evidence, and actionable recommendations
Answer business questions from read-only connected databases using deterministic schema discovery and SQL evidence
Run Python code in an isolated subprocess for statistical/ML/visualization work beyond what SQL can express
End-to-end workflow for adding a new pipeline node to a seeknal project — scaffold, validate, apply, and (optionally) run via the 5 thin pipeline-build tools
| name | complex-analysis |
| description | Run multi-step SQL plus Python/statistics/ML analysis while keeping tools thin and evidence grounded |
| tags | ["complex-analysis","python","machine-learning","statistics"] |
| version | 1.0.0 |
Use this workflow when the user asks for deeper analysis than a single SQL aggregation: correlations, clustering, forecasting, simple machine-learning models, anomaly detection, visualization, or custom scoring.
Scope with SQL first
list_tables/describe_table when schema is unknown.execute_sql to confirm row counts, columns, and a small preview.Run Python only when it adds value
execute-python-analysis before execute_python.conn object. Do not import duckdb or create a new connection.conn.sql("SELECT ...").df().pd),
numpy (np), matplotlib (plt), scikit-learn (sklearn), and scipy
(scipy). Check optional libraries before using them; if plotting is
unavailable, return text/table evidence instead of retrying chart code.Model responsibly
Return evidence
Connected source access remains read-only. Do not mutate external databases, write pipeline files, or publish artifacts unless the user explicitly asks and the relevant approval-gated skill allows it.