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.
git clone --depth 1 https://github.com/skillmds/skillmdWrote 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/plugins/skillmds/skillmd/data-ml)<a href="https://agentmods.dev/plugins/skillmds/skillmd/data-ml"><img src="https://agentmods.dev/badge/plugins/skillmds/skillmd/data-ml/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/plugins/skillmds/skillmd/data-ml"><img src="https://agentmods.dev/badge/plugins/skillmds/skillmd/data-ml.svg" alt="Reviewed on agentmods" width="80" height="20"></a>Grade A, and why
data-ml 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 4d 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.
What it actually says
{
"name": "data-ml",
"description": "SQL, analytics, datasets, models and machine-learning workflows.",
"version": "1.0.0",
"author": {
"name": "SkillMD"
},
"homepage": "https://skillmd.com/plugins/skillmd/data-ml"
}
What it installs
The manifest is a name and a version. 17 skills travel with it, and installing the plugin installs all of them — 1,343 tokens a session between them. Each is measured on its own page, and each can be installed alone.
- Skill dbs-diagnosis A 126 tokens
- Skill pandera-polars A 112 tokens
- Skill pyarrow-python A 117 tokens
- Skill diagram A 63 tokens
- Skill ktx A 81 tokens
- Skill Writingmate MCP Video and Image Generation A 58 tokens
- Skill Claude-design-analysis A 123 tokens
- Skill weather-plugin A 23 tokens
- Skill ml A 5 tokens
- Skill agentic-kaggle-skill A 110 tokens
- Skill surrealdb A 204 tokens
- Skill hqq-quantization A 58 tokens
- Skill rag A 44 tokens
- Skill sql-pro A 36 tokens
- Skill captions-overlay A 136 tokens
- Skill data-report A 19 tokens
- Skill flowai-team-dashboard A 28 tokens
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.
- 4d ago First seen · 10 lines scan A a481bf60a73f
data-ml is a plugin published in the GitHub repository skillmds/skillmd (1 stars, last pushed yesterday), licensed MIT. Its token cost is not measured: this kind of file is read by the harness, not the model. 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 2026-09-19.
Other plugins, from other repositories
scraps-migration
LLM migration workflows for moving Scraps wikis from v0 to v1. Runs version-pinned audits, converts tag and config syntax, and verifies the result with v1 lint/build.
poly
Zero-dependency lint & format orchestrator over MCP — one pure-Rust binary wrapping ruff/oxc/taplo/rumdl and more as in-process backends, plus a tree-sitter generic tier for everything else. Exposes lint/format/cache tools with structured JSON output.
scomp-link
End-to-end machine learning toolkit with 15 MCP tools. Train models, tune hyperparameters, detect drift, generate HTML reports with 39 chart types, detect anomalies, forecast time series, check fairness, and serve models.
ensemble-ai
AI SDK skills for OpenAI, and other AI providers for Claude Code.
data-team
A data-work team covering data collection and cleanup, analysis, and visualisation. Visualisation means turning data into charts or other graphical explanations.
estelle
Estelle — code memory + the grounding gate, always on. The trust layer under your AI coding agents: it grounds every model in your real codebase and refuses to ship what isn't true.