ds-reviewer

ds-reviewer is an agent for Claude Code from StamKavid/last-ds-mile. It costs 70 tokens per session (466 once invoked), scanned A, original, MIT.

A reviewer for data-science notebooks and pipelines, using a checklist before results are reported or handed over. A pipeline is the sequence of steps that prepares data, trains a model, and measures it.

In plain words
What is it for?
Use it to check whether a project has a simple comparison model, an appropriate validation split, a suitable metric, and subgroup performance checks.
Why use it?
It exposes missing baselines, unsuitable testing methods, mismatched metrics, and poor performance hidden in particular groups.

Agent for Claude Code

Written for Claude Code: effort in frontmatter. Also seen: model in frontmatter.

Part of the last-ds-mile plugin — 29 skills, 17 commands, 3 agents, 4 hooks shipped together

Good fit Use it to check whether a project has a simple comparison model, an appropriate validation split, a suitable metric, and subgroup performance checks.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/stamkavid/last-ds-mile/ds-reviewer
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.

Clone the repo
git clone --depth 1 https://github.com/StamKavid/last-ds-mile

Made for: Claude Code.

Or install last-ds-mile, the plugin that ships this one along with the rest of its 29 skills, 17 commands, 3 agents, 4 hooks.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for ds-reviewer

README.md
[![agentmods](https://agentmods.dev/badge/agents/stamkavid/last-ds-mile/ds-reviewer/github.svg)](https://agentmods.dev/agents/stamkavid/last-ds-mile/ds-reviewer)
Your own site
<a href="https://agentmods.dev/agents/stamkavid/last-ds-mile/ds-reviewer"><img src="https://agentmods.dev/badge/agents/stamkavid/last-ds-mile/ds-reviewer/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for ds-reviewer

Your own site · 80×15
<a href="https://agentmods.dev/agents/stamkavid/last-ds-mile/ds-reviewer"><img src="https://agentmods.dev/badge/agents/stamkavid/last-ds-mile/ds-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 70 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 466 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00070 $0.00466
Opus 5 $0.00035 $0.00233
Sonnet 5 $0.00014 $0.00093
Haiku 4.5 $0.00007 $0.00047

Measured 10d ago against content hash e1523370b514, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

ds-reviewer 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 10d 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.

agents/ds-reviewer.md · 22 lines

What it actually says

You are a data science pipeline reviewer applying the last-ds-mile plugin's discipline checklist (see the ds-method skill for the canonical list) to a notebook, script, or set of .last-ds-mile/stages/*.md files before it's reported or handed off.

Check, in order, and cite the specific evidence (or its absence) for each:

  1. Honest baseline — is there a documented naive/simple baseline (majority class, mean predictor, or simplest reasonable model) that the final model is compared against? A model without a baseline comparison is an unverified claim of value.
  2. Validation strategy — does the validation approach match the data's structure (time-series split for temporal data, group-aware split for grouped data, stratification for imbalanced targets)? Is the same validation scheme used consistently, not swapped mid-pipeline?
  3. Metric selection — does the chosen metric match the actual problem (e.g. accuracy on an imbalanced target is a Red Flag; is precision/recall/F1/AUC-PR considered)?
  4. Slice performance — has performance been checked on meaningful subgroups, not just in aggregate? An aggregate score can hide a model that fails badly on a subpopulation.
  5. Reproducibility — are random seeds set, package versions recorded, and is the pipeline runnable end to end without manual steps that aren't documented?

For each of the 5 checks, report PASS (with the evidence) or FAIL (with exactly what's missing and where to add it). Do not soften a FAIL into a suggestion — these are the hard gates the plugin's stages are built around.

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. 10d ago First seen · 22 lines · 70 tokens per session scan A e1523370b514

Subscribe to this mod's changes

ds-reviewer is an agent published in the GitHub repository StamKavid/last-ds-mile (3 stars, last pushed 1mo ago), licensed MIT. It adds 70 tokens to every session and 466 once invoked, about $0.0003 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.

Related

Other agents, from other repositories

senior-data-scientist

Reviews statistical analyses, ML pipelines, data quality, model validation, and data serialization practices.

pjt222/agent-almanac · 24 tokens

part-sourcer

Use this agent to find the best JLCPCB-stocked LCSC part matching a generic component spec. Given something like "3.3V LDO regulator, SOT-23-5, 500mA out, basic-tier preferred", returns one recommended pick plus 2 alternates with cost-impact annotations. Hard-prefers basic-library parts to avoid the JLCPCB…

BeckhamLabsLLC/kicad-jlcpcb · 198 tokens

Demonstrate

Agent for demonstrating VS Code features.

microsoft/vscode · 10 tokens

playwright-test-generator

Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.

microsoft/playwright · 151 tokens

AVM Owner Triage

Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.

github/awesome-copilot · 61 tokens

Ultimate Transparent Thinking Beast Mode

Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.

github/awesome-copilot · 11 tokens