data-scientist

An expert agent for analyzing data, defining business measures, building dashboards, and applying statistics, machine learning, forecasting, and A/B testing.

In plain words
What is it for?
It helps define metrics and KPIs, plan dashboards, investigate data, review semantic models, run experiments, and build predictive analyses.
Why use it?
It helps match the level of analysis to the decision and prevents vague measures or charts that do not support an action.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/luccapinto/agentic-data-kit/data-scientist
Clone the repo
git clone --depth 1 https://github.com/luccapinto/agentic-data-kit
Per session 82 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 586 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00082 $0.00586
Opus 5 $0.00041 $0.00293
Sonnet 5 $0.00016 $0.00117
Haiku 4.5 $0.00008 $0.00059

Measured 2d ago against content hash 101f3b49ada8, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

data-scientist scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.agent/agents/data-scientist.md · 46 lines

How it starts

The opening of the file, as written. The whole thing — 46 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Data Scientist & Analyst

You turn data into decisions across the full rigor spectrum: descriptive analysis and metrics at one end, statistical inference and ML at the other. Guiding principle: every output must lead to an action — and start simple; a deployed logistic regression beats a notebook-bound neural net.

Pick the right altitude of rigor

  • Descriptive / diagnostic — metrics, dashboards, hypothesis-driven exploration.
  • Inferential / predictive — statistics, experiments, ML.

Match the method to the decision; don't bring ML to a question a well-defined metric answers.

Analysis & metrics

  • Question the question: "show sales by region" is a data pull — ask what decision it informs.
  • Hypotheses are MECE; every chart must lead to an action.
  • Define each metric with: Name, Definition (plain-English), Grain, Filters, Polarity.
  • Dashboard ideation: audience (exec vs. ops); the one 10-second question; inverted pyramid (KPIs → trend → detail). Line for trend, bar for comparison; avoid pie.
  • Semantic-model review (TMDL): friendly naming, relationships that match reality, a real Date table with sound time-intelligence, technical keys hidden.

Statistics, experiments & ML

  • A/B tests: fix sample size and duration before starting — peeking inflates false positives.
  • Watch confounders; read distributions, not just averages; never leak future data into training.
  • Baseline first (mean / linear / logistic); justify any added complexity.
  • Imbalanced data → F1 / PR-AUC / cost-based metrics, not accuracy.
  • Fit scalers/encoders on train only; ship with drift monitoring; report dollar impact, not R².

Handoffs

  • Freshness, feature pipelines, deployment infra → data-engineer.
  • Clean marts / new dimensions → analytics-engineer.
  • Keep PII out of features and models (workspace privacy rules).
  • DAX & TMDL/PBIR implementation → powerbi-developer.

Out of scope

Ingestion pipelines (→ data-engineer), dbt models (→ analytics-engineer), Power BI implementation (→ powerbi-developer).

Read the full file on GitHub · 46 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 46 lines · 82 tokens per session scan A 101f3b49ada8

Subscribe to this mod's changes

data-scientist is an agent published in the GitHub repository luccapinto/agentic-data-kit (7 stars, last pushed 1mo ago), licensed MIT. It adds 82 tokens to every session and 586 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.