seeknal

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

Run multi-step SQL plus Python/statistics/ML analysis while keeping tools thin and evidence grounded

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

Conversational data ingestion — parse files, brainstorm schema, create queryable tables, and save reusable skills

Data-entry, assistant, and advisor: record one-off events from text (/record ...) or receipt/transfer images into the right ingest table, asking questions until the row is complete

Auto-generate draft semantic model YAML from CSVs and parquets by profiling columns into entities/dimensions/measures

Apply a full-document rewrite to an existing Proof Editor document via the rewrite.apply op (CAUTION: disruptive, rejected when live collaborators are present)

Profile CSV files in data/ for schema, quality, null counts, unique counts, sample values, and join-key candidates

Publish a markdown memo to a Proof Editor server (memokami.exe.xyz / proofeditor.ai / self-hosted) and return a shareable URL

Publish a built Evidence.dev report to a Seeknal Report Server and return a shareable URL

Query business metrics through the semantic layer with automatic joins, time grain resolution, and alias resolution

End-to-end Evidence.dev report workflow — exploration, drafting, approval-gated build, codification, and post-report next-steps menu

Codify an ad-hoc metric query as a permanent seeknal/metrics/{name}.yml definition reusable via query_metric

Codify a completed analysis as a repeatable seeknal/exposures/{name}.yml spec re-runnable via `seeknal ask report --exposure {name}`

Run spec-driven QA automation for seeknal medallion E2E pipelines. Discovers YAML specs in qa/specs/, spawns parallel worker agents per spec, scaffolds seeknal projects, executes against live infrastructure (CSV, Iceberg, PostgreSQL), and validates results end-to-end. Inspired by Bowser's composable automation pattern.