Skip to content

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

130 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Assay

Verified statistical research on tabular datasets. Point it at a CSV or Kaggle slug and it runs a full research loop — with the constraint that every number the LLM reports is independently recomputed by a deterministic verifier before it reaches the report.

How it works

  1. Profile — deterministic dataset profile + data-quality scan (no LLM involved)
  2. Plan — an LLM proposes research questions against the profile; you approve/edit them at a gate (CLI or app)
  3. Explore — analysis code runs in a live Jupyter kernel
  4. Verify — every reported statistic is recomputed from canonical formulas by verifier.py (Welch, ANOVA, Pearson/Spearman, chi², Mann-Whitney, Kruskal, linear/logistic regression, grouped variants) and marked confirmed / discrepancy / unverifiable. Benjamini-Hochberg correction is applied once across all rounds, with an effect-size floor.
  5. Follow up — a findings-informed second round, gated again
  6. Publish — journal-style report (95% CIs, robustness lines, data-quality section, house-style charts) + a cold-run-verified notebook that re-runs on Kaggle as-is, with optional gated Kaggle publishing (private by default) and portfolio export

Interfaces

  • CLI: python run.py <csv-or-kaggle-slug>
  • App: Electron + React three-panel UI over a stdio JSON protocol — gates pause the session for review, reports are editable in-place, old runs can be re-rendered

Credentials

Nothing is hardcoded. Kaggle credentials come from KAGGLE_USERNAME/KAGGLE_KEY env vars (or kaggle.json); the LLM key loads via env var. Publishing to Kaggle requires an explicit identity confirmation gate.

Tests

~220 across the Python engine and UI. The verifier is exercised against known-answer statistical fixtures.

Status

Working and in active use — reports and figures from real runs feed the research section of my portfolio. No license; all rights reserved.

About

Verified statistical research assistant — LLM plans and explores, a deterministic verifier independently recomputes every reported number before it reaches the report

Topics

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages