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-iterategit 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-iterate)<a href="https://agentmods.dev/skills/stamkavid/last-ds-mile/ds-iterate"><img src="https://agentmods.dev/badge/skills/stamkavid/last-ds-mile/ds-iterate/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-iterate"><img src="https://agentmods.dev/badge/skills/stamkavid/last-ds-mile/ds-iterate.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 | $0.00081 | $0.01377 |
| Opus 5 | $0.00041 | $0.00688 |
| Sonnet 5 | $0.00016 | $0.00275 |
| Haiku 4.5 | $0.00008 | $0.00138 |
Grade A, and why
ds-iterate 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.
How it starts
The opening of the file, as written. The whole thing — 87 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ds-iterate — Diagnose and Route Back
Overview
/ds-frame through /ds-handoff reads like a straight line, but real modeling work is
a loop: evaluate, find what's wrong, fix the specific thing, re-evaluate. This stage is
the diagnosis-and-routing step that turns one pass into a real iteration, instead of
running the pipeline once and calling it done despite a fixable weakness in
/ds-evaluate's own error analysis.
When to Use
- Immediately after
/ds-evaluate, before deciding whether to proceed to/ds-explainor go back for another pass. - The aggregate metric is acceptable but a slice, an error-analysis pattern, or a
calibration issue from
/ds-evaluatesuggests a fixable, specific weakness. - NOT for: re-running the exact same modeling step hoping for a better random draw
(that's not iteration, that's noise-chasing — see
uncertainty-quantificationfor whether a gap is even real) — this stage requires a specific, named fix.
Core Process
- Read
.last-ds-mile/stages/07-evaluate.mdin full: the slice table, the calibration check, and the worst-mispredictions pattern. Do not skip straight to a verdict — the diagnosis has to come from what's actually written there. - Categorize the gap using the table below. Pick the category the evidence actually supports, not the one that's easiest to act on.
- If the diagnosis points to bias (systematic underperformance everywhere, including on training data) or variance (train much better than validation), check for a learning-curve signal before deciding the fix: does more data help (variance), or does a more expressive model/feature set help (bias)? State which, briefly.
- Route back to the one prior stage that addresses the diagnosed cause — not a generic "try again." Re-run only that stage; don't restart the whole pipeline.
- Cap iteration: after 3 loops on the same problem without the diagnosed issue
resolving, stop looping and say so plainly — report the unresolved gap as a known
limitation for
/ds-reportrather than iterating indefinitely chasing a better number. - Append one entry to
.last-ds-mile/stages/07-iterate-log.mdper loop: the diagnosis, the stage routed back to, what changed, and the resulting metric versus the previous loop's — so the loop's history is auditable, not silently overwritten.
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.
- 10d ago First seen · 87 lines · 81 tokens per session scan A ead9640de86d
ds-iterate is a skill published in the GitHub repository StamKavid/last-ds-mile (3 stars, last pushed 1mo ago), licensed MIT. It adds 81 tokens to every session and 1,377 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.