tkpratardan

tkpratardan/lemma

Plugin Claude Code

Senior data-scientist mode + live-notebook tools for AI agents.

tkpratardan/lemma

Plugin Claude Code

Senior data-scientist mode for AI agents, plus live-notebook tools. Frame the question, look before you model, baseline before complexity, validate honestly.

tkpratardan/lemma

Cursor rule Cursor

Lemma senior data-scientist mode for data/notebook work.

tkpratardan/lemma

Cursor rule Cursor

Use Lemma's canonical notebook edit action and honor editor approval.

tkpratardan/lemma

Instructions file GitHub Copilot

Instructions for tkpratardan/lemma: Act as a senior data scientist. Answer from executed evidence in the active notebook.

tkpratardan/lemma

MCP server Claude CodeCodexCursor +2

MCP server "lemma" as configured in tkpratardan/lemma. Runs ${CLAUDE_PLUGIN_ROOT}/bin/lemma-mcp.mjs with node.

tkpratardan/lemma

Skill Claude CodeCodex

Establish a dumb baseline and an honest validation harness before any real model, so every later number means something.

tkpratardan/lemma

Skill Claude CodeCodex

Rigor for causal questions and A/B tests (the effect of acting on X): confounding, post-treatment bias, valid control groups.

tkpratardan/lemma

Skill Claude CodeCodex

Rigor for descriptive and diagnostic analytics (what happened and why): denominators, grain, and confounded slices, not model leakage.

tkpratardan/lemma

Skill Claude CodeCodex

EDA kickoff for a fresh dataset: fixed opening scaffold (goal, imports, load, sanity), then chapters derived from the data; scan leakage, land a baseline.

tkpratardan/lemma

Skill Claude CodeCodex

Rigor for statistical inference (is the difference real): hypothesis tests, power, multiple comparisons, effect size over p-value.

tkpratardan/lemma

Skill Claude CodeCodex

Audit a dataset or pipeline for the five leakages that inflate a metric: target, preprocessing, temporal, group, and sampling.

tkpratardan/lemma

Skill Claude CodeCodex

Final modeling once the baseline and feature set are locked: tune against validation, audit overfitting, touch the test set once, justify the complexity.

tkpratardan/lemma

Skill Claude CodeCodex

Review a notebook or analysis for data-science anti-patterns before it's trusted or shared.

tkpratardan/lemma

Skill Claude CodeCodex

Rigor for clustering, dimensionality reduction, and anomaly detection: validity is stability under resampling, not a held-out score.

tkpratardan/lemma

Skill Claude CodeCodex

Assemble a trustworthy working dataset from messy or multiple sources: grain, keys, joins with match rates, extraction checks, lineage.

tkpratardan/lemma

Instructions file CodexOpenCode

Instructions for tkpratardan/lemma: Act as a senior data scientist. Answer from executed evidence in the active notebook.

tkpratardan/lemma

Hook

Runs when a session starts on startup, resume, clear and compact, executing session-start.js via node. From tkpratardan/lemma.

tkpratardan/lemma

Hook

Runs when you submit a prompt, before the agent sees it, executing prompt-submit.js via node. From tkpratardan/lemma.

tkpratardan/lemma

Hook

Runs after a tool call finishes for mcp__(lemma and plugin_lemma_lemma)__.* tool calls, executing post-tool-use.js via node. From tkpratardan/lemma.

tkpratardan/lemma

Hook

Runs before the agent uses a tool, executing pre-tool-use.js via node. From tkpratardan/lemma.

tkpratardan/lemma

Hook

Runs when the agent finishes a response, executing stop.js via node. From tkpratardan/lemma.

tkpratardan/lemma

Hook

Runs when a subagent starts, executing subagent-start.js via node. From tkpratardan/lemma.