stas00/ml-engineering

DevOps & infra
Coding workflow

Graded C on context cost and craft. 440 tokens of this skill ride in the system prompt of every message, whether or not it fires.

Ranked #197 of 238 on the leaderboard →

Score: 1.90 / 4
Resident: 440 tokens
Footprint: medium
Per activation: 25,805 tokens

Share this grade on X ↗Re-grade now →View on GitHub ↗Grade another skillLast graded Nov 2, 2026 from HEAD

Add the badge to your README

It shows the current letter grade and links back to this page. Improve the skill, hit Re-grade now, and the badge follows — GitHub may keep the old image for up to an hour.

Markdown

[![Skill Grader: C](https://seoagent.com/skill-grader/badge/stas00/ml-engineering.svg)](https://seoagent.com/skill-grader/stas00/ml-engineering)

HTML

<a href="https://seoagent.com/skill-grader/stas00/ml-engineering"><img src="https://seoagent.com/skill-grader/badge/stas00/ml-engineering.svg" alt="Skill Grader: C" height="20"></a>

Report card

C overall · 1.90/4

github.com/stas00/ml-engineering@HEAD · 3 skills · 440t resident (medium) · 25,805t per activation · 401,985t bundle

Resident footprint

440t of description text ride in the system prompt of every message across 3 skills (paper's observed per-skill range: 50–280t).

Description honesty

Defensive padding detected: 23 comma-separated trigger keywords; 210 words (a quiet description is ~15–40). Padding is individually rational and collectively ruinous — it dilutes every co-installed skill.

Body size

Largest body is 12,755 words (~25,805t), loaded on every trigger. Corpus median 921, p90 2207.

Progressive disclosure

59 auxiliary .md files ship but the body never points at them — the agent won't find them.

Factoring

Scoped to a coherent procedure set.

CLI leverage

Drives a CLI (`dcgmi`, 8 invocations + 16 shell blocks) — verification and state live in code, not prompt.

C overall (1.90/4). 3 skill(s), 440t resident on every message (medium), 25,805t loaded per activation. Weakest axis: body size (F).

Six deterministic axes, no LLM, same result every run. Tokens are approximated as chars/4. Read the methodology, or see how SEO skills score on task outcomes in the open SEO skill benchmark.