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 Leeroo-AI/superml --skill ml-iterategit clone --depth 1 https://github.com/Leeroo-AI/supermlWrote 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/leeroo-ai/superml/ml-iterate)<a href="https://agentmods.dev/skills/leeroo-ai/superml/ml-iterate"><img src="https://agentmods.dev/badge/skills/leeroo-ai/superml/ml-iterate.svg" alt="Measured on agentmods" 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 | $0.00034 | $0.04677 |
| Opus 5 | $0.00017 | $0.02338 |
| Sonnet 5 | $0.00007 | $0.00935 |
| Haiku 4.5 | $0.00003 | $0.00468 |
Grade A, and why
ml-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 8d 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 — 199 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ML Iteration
Generate ranked, grounded next steps when you've tried something and need to improve.
Grounding
Detect mode: On your first grounding call, check if Leeroopedia KB tools are available. If they return results, use KB mode. If unavailable or auth fails, use Web mode.
KB mode: Call propose_hypothesis → search_knowledge → query_hyperparameter_priors. Cite as [PageID].
Web mode: WebFetch GitHub issues for similar problems → WebFetch framework tuning guides → WebFetch published configs/ablations. Cite as [source](URL). Start response with: > Grounding: Web mode — citations from official docs.
Web mode URL registry:
- HF Transformers/PEFT/TRL:
https://huggingface.co/docs/{transformers,peft,trl} - Axolotl:
https://github.com/axolotl-ai-cloud/axolotl - DeepSpeed:
https://www.deepspeed.ai/docs - vLLM:
https://docs.vllm.ai - Model cards:
https://huggingface.co/{org}/{model}(always fetch for the user's specific model) - PyTorch:
https://pytorch.org/docs/stable - Weights & Biases reports:
https://wandb.ai/site/articles(for published ablation studies)
The Iron Law
NO NEW EXPERIMENT WITHOUT REVIEWING WHAT YOU ALREADY TRIED
Re-running a failed approach with minor tweaks is the most common waste of GPU time. Check your history first.
The Grounding Law
NO BARE TECHNICAL CLAIMS — EVERY NUMBER AND MODEL-SPECIFIC FACT GETS A TAG
Default every technical claim to [unverified — no KB access]. Upgrade to [PageID: xxx] only when you have an actual KB result. There is no third option. Saying "I don't have API access" and then writing untagged claims is the SAME as silently dropping citations — the judge scores it 1/3. Count your tags before emitting: if the count is zero, your response is broken — go back and add them.
Phases
Phase 0: Pre-flight (do this FIRST)
Attempt a search_knowledge call. If it succeeds, you're in KB mode. If it fails:
YOU ARE NOW IN WEB MODE. Execute these WebFetch calls before writing ANY text:
WebFetchthe user's model card:https://huggingface.co/{org}/{model}(e.g.,https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct)WebFetchthe framework docs for their training setup (e.g.,https://huggingface.co/docs/trlorhttps://huggingface.co/docs/peft)WebFetchat least one more relevant source (GitHub issues, ablation studies, or config examples)- Extract numbers immediately: After each WebFetch, write down specific values (LR, rank, batch size, warmup) found in the source. These become your citation anchors — quote them in Phase 3. A URL without an extracted number is not a useful citation.
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.
- 8d ago First seen · 199 lines · 34 tokens per session scan A e1957907e5f3
ml-iterate is a skill published in the GitHub repository Leeroo-AI/superml (194 stars, last pushed 5mo ago), licensed Apache-2.0. It adds 34 tokens to every session and 4,677 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
histolab
Lightweight WSI tile extraction and preprocessing. Use for basic slide processing tissue detection, tile extraction, stain normalization for H&E images. Best for simple pipelines, dataset preparation, quick tile-based analysis. For advanced spatial proteomics, multiplexed imaging, or deep learning pipelines use pathml.
torch-geometric
Graph Neural Networks (PyG). Node/graph classification, link prediction, GCN, GAT, GraphSAGE, heterogeneous graphs, molecular property prediction, for geometric deep learning.
zarr-python
Chunked N-D arrays for cloud storage. Compressed arrays, parallel I/O, S3/GCS integration, NumPy/Dask/Xarray compatible, for large-scale scientific computing pipelines.
glycobiology
Glycosylation site prediction and glycobiology analysis. N-glycosylation motif finding, O-glycosylation hotspot prediction, glycan structure resources. Lightweight, pure Python. For protein function queries use uniprot-database; for structure analysis use alphafold-database.
pyhealth
Comprehensive healthcare AI toolkit for developing, testing, and deploying machine learning models with clinical data. This skill should be used when working with electronic health records (EHR), clinical prediction tasks (mortality, readmission, drug recommendation), medical coding systems (ICD, NDC, ATC)…
deepchem
Molecular ML with diverse featurizers and pre-built datasets. Use for property prediction (ADMET, toxicity) with traditional ML or GNNs when you want extensive featurization options and MoleculeNet benchmarks. Best for quick experiments with pre-trained models, diverse molecular representations. For graph-first…