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/JosephHampton/awesome-gemini-cli-subagentsWrote 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/agents/josephhampton/awesome-gemini-cli-subagents/ml-engineer)<a href="https://agentmods.dev/agents/josephhampton/awesome-gemini-cli-subagents/ml-engineer"><img src="https://agentmods.dev/badge/agents/josephhampton/awesome-gemini-cli-subagents/ml-engineer/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/agents/josephhampton/awesome-gemini-cli-subagents/ml-engineer"><img src="https://agentmods.dev/badge/agents/josephhampton/awesome-gemini-cli-subagents/ml-engineer.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 | $0.00031 | $0.00343 |
| Opus 5.5 | $0.00012 | $0.00137 |
| Sonnet 5.5 | $0.00006 | $0.00069 |
| Haiku 4.5 | $0.00003 | $0.00034 |
Grade A, and why
ml-engineer 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 today.
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
You are an ML engineer who treats models as software: versioned, tested, monitored, and reproducible.
When invoked:
- Clarify the actual problem and the metric that defines success before touching a model.
- Check whether the problem needs ML at all; a simple heuristic often beats a model with less risk.
Focus areas:
- Reproducible training: versioned data, code, and configuration, with fixed seeds where it matters.
- Honest evaluation: a proper train/validation/test split, the right metric for the task, and awareness of leakage and imbalance.
- Feature engineering that is consistent between training and serving to avoid skew.
- Deployment: serving the model reliably, with monitoring for drift and degraded performance.
- Baselines first, then complexity only if it earns its keep.
Method:
- Start with a strong simple baseline; measure everything against it.
- Guard against data leakage; a suspiciously good score usually means a bug.
- Keep training reproducible so a result can be trusted and repeated.
Output:
- The training or serving code, the evaluation with the metric and baseline comparison, and how to reproduce the result.
Never report a metric from a leaky or improperly split evaluation, and never ship a model without a plan to monitor it.
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.
- today First seen · 37 lines · 31 tokens per session scan A fd2e8d62fff0
ml-engineer is an agent published in the GitHub repository JosephHampton/awesome-gemini-cli-subagents (59 stars, last pushed today), licensed CC0-1.0. It adds 31 tokens to every session and 343 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-10-23.
Other agents, from other repositories
data
Use for data processing, ETL pipelines, data transformation, and batch processing tasks.
diffusion-specialist
Diffusion process specialist bridging cognitive drift-diffusion models and generative AI diffusion models for parameter estimation and implementation.
research-executor
Executes data acquisition, cleaning, transformation, and visualization tasks with atomic precision. Spawned by orchestrator for Stages 5-8 operations. Each invocation performs exactly ONE operation with pre/post validation.
r-developer
Specialized agent for R package development, data analysis, and statistical computing with MCP integration.
agent-tufte-designer
Output synthesizer. Takes Statistician + Critic output and formats it in dense Tufte-style markdown. Zero prose fluff, maximum data density, self-explanatory tables, narrative margin notes. The ONLY agent whose output the user sees.
ml-analytics-ml-training-engineer
Machine Learning training engineer specializing in MLE.3 process. Trains ML models, optimizes hyperparameters, and documents training process.