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.
npx skills add StamKavid/last-ds-mile --skill ds-baselinegit clone --depth 1 https://github.com/StamKavid/last-ds-mileWrote 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.
[](https://agentmods.dev/skills/stamkavid/last-ds-mile/ds-baseline)<a href="https://agentmods.dev/skills/stamkavid/last-ds-mile/ds-baseline"><img src="https://agentmods.dev/badge/skills/stamkavid/last-ds-mile/ds-baseline/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.
<a href="https://agentmods.dev/skills/stamkavid/last-ds-mile/ds-baseline"><img src="https://agentmods.dev/badge/skills/stamkavid/last-ds-mile/ds-baseline.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00075 | $0.00773 |
| Opus 5 | $0.00037 | $0.00387 |
| Sonnet 5 | $0.00015 | $0.00155 |
| Haiku 4.5 | $0.00007 | $0.00077 |
Grade A, and why
ds-baseline 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 11d 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.
How it starts
The opening of the file, as written. The whole thing — 64 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ds-baseline — Honest Baseline
Overview
Establishes the dumbest reasonable prediction as the anchor metric, so every later model's score can be judged as real lift or noise, not judged in a vacuum.
When to Use
- Before
/ds-model— this is a Hard Gate/ds-modelchecks for. - Whenever asked to build, train, or compare models and no baseline artifact exists yet for this problem.
- NOT for: tuning or comparing real candidate models (that's
/ds-model) — this stage produces exactly one deliberately simple number to compare against. - Scope: tabular supervised learning. On a time-indexed target the right anchor is seasonal-naive or last-value, and this plugin does not carry the lag/rolling feature machinery to go further — see README → Scope before using this as a forecasting baseline.
Core Process
- Pick the simplest possible baseline for the problem type: mean or median prediction for regression; majority-class or prior-probability prediction for classification; or a simple rule already in informal use, if one exists. Simplest still has to be honest: on data with strong known structure (temporal, seasonal, hierarchical), a global mean or majority class is a strawman — trivial to beat, so beating it proves nothing. Use the strongest simple anchor that needs no modeling: last known value, same period last cycle (e.g. same weekday last week), or the rule already in use.
- Evaluate it using the exact success metric chosen in
/ds-frame— not a different, more convenient metric. - Record the baseline score as the anchor. Every subsequent model must be compared against it, not against zero or against "feels better."
- Write to
.last-ds-mile/stages/04-baseline.md: the baseline definition, its score, and what "beating it" will concretely mean.
Common Rationalizations
| Rationalization | Reality |
|---|---|
| "This is a well-known dataset/problem, everyone knows a baseline would be trivial" | Trivial to state is not the same as trivial to skip — it's the only thing that tells you whether your fancier model earned its added complexity. |
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.
- 11d ago First seen · 64 lines · 75 tokens per session scan A f6fdf18201d4
ds-baseline is a skill published in the GitHub repository StamKavid/last-ds-mile (3 stars, last pushed 1mo ago), licensed MIT. It adds 75 tokens to every session and 773 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.
Other skills, from other repositories
marimo-pair
Work inside the user's live marimo notebook from the code editor: run Python in the same kernel the user does, inspect live notebook state, and commit durable notebook changes through code mode. Use whenever you create, analyze, or improve the user's marimo notebook.
fill-model-descriptions
Fill missing and refresh obsolete model descriptions in packages/llm-info/data/models.yml by querying OpenRouter and provider documentation. Use when the user asks to populate model descriptions, enrich the model catalog, or curate descriptions after running pnpm sync-models.
ml-expert
Expert-level machine learning, deep learning, model training, and MLOps. Use when the user mentions machine learning, deep learning, neural networks, MLOps, or data science, or when the task involves Machine Learning Fundamentals, Data Preparation, or Model Training.
scomp-link
End-to-end ML toolkit with 26 CLI commands. Use when training models, tuning hyperparameters, detecting data drift, generating HTML reports with charts, profiling datasets, detecting anomalies, forecasting time series, checking fairness, or serving models as REST APIs. Prefer over raw sklearn when you need automated…
agentic-data-science-competition
AI Agent-driven Kaggle competition workflow. Learn from real competition experience: score stabilization patterns, submission troubleshooting, kernel workflows, GPU task delegation, and the spec-driven development approach that achieved top leaderboard positions. Use when: working on any Kaggle competition, analyzing…
data-analysis
Structured data analysis workflow from raw data to shareable insights.