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-handoffgit 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-handoff)<a href="https://agentmods.dev/skills/stamkavid/last-ds-mile/ds-handoff"><img src="https://agentmods.dev/badge/skills/stamkavid/last-ds-mile/ds-handoff/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-handoff"><img src="https://agentmods.dev/badge/skills/stamkavid/last-ds-mile/ds-handoff.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 | $0.00070 | $0.00611 |
| Opus 5 | $0.00035 | $0.00305 |
| Sonnet 5 | $0.00014 | $0.00122 |
| Haiku 4.5 | $0.00007 | $0.00061 |
Grade A, and why
ds-handoff 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 12d 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 — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ds-handoff — Reproducibility & Handoff
Overview
Packages the finished work so someone else can rerun it, trust it, and know its limits — the last mile the whole plugin is named for.
When to Use
- The model or analysis is ready to be shared, deployed, or handed to another team.
- After
/ds-reporthas produced the stakeholder narrative and limitations. - NOT for: further tuning or re-evaluation — if the model isn't finished, go back to
/ds-modelor/ds-evaluatefirst.
Core Process
- Gate check: confirm the environment is pinned — a lockfile, or a
requirements.txt/environment.ymlwith exact versions, not bare package names. If it isn't pinned, stop and pin it before continuing. - Write a model card: what it predicts, a training-data summary, the metric and its
baseline lift, known limitations (from
/ds-report), and intended vs. out-of-scope use. - Confirm the notebook or script reruns cleanly top-to-bottom in a fresh kernel/environment before considering the work done.
- Serialize the model artifact together with its version and the training-data hash/date, recorded alongside it.
- Write to
.last-ds-mile/stages/10-handoff.md: the model card, the environment pin location, the rerun confirmation, and the artifact location.
Common Rationalizations
See ds-method for the shared Rationalizations that apply to every stage, in particular
"it's just exploratory, I'll clean this up before shipping" — the notebook-nobody-could-
rerun failure mode this stage exists to prevent.
Red Flags
| Red Flag | What it usually means |
|---|---|
| The environment file lists packages with no version numbers | This is exactly the condition the Gate Check above exists to catch — unpinned environments are how "works on my machine" ships. |
See ds-method for the shared Red Flags that apply to every stage.
See lessons/the-notebook-nobody-could-rerun.md for a real example of a
handoff that looked done but wasn't reproducible.
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.
- 12d ago First seen · 58 lines · 70 tokens per session scan A 209ef27459fd
ds-handoff is a skill 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 611 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.
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.