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 SFETNI/Deep-Matter-Chem-Skills --skill fine-tuning-mlpgit clone --depth 1 https://github.com/SFETNI/Deep-Matter-Chem-SkillsWrote 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/sfetni/deep-matter-chem-skills/fine-tuning-mlp)<a href="https://agentmods.dev/skills/sfetni/deep-matter-chem-skills/fine-tuning-mlp"><img src="https://agentmods.dev/badge/skills/sfetni/deep-matter-chem-skills/fine-tuning-mlp/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/sfetni/deep-matter-chem-skills/fine-tuning-mlp"><img src="https://agentmods.dev/badge/skills/sfetni/deep-matter-chem-skills/fine-tuning-mlp.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 | $0.00006 | $0.14205 |
| Opus 5 | $0.00003 | $0.07103 |
| Sonnet 5 | $0.00001 | $0.02841 |
| Haiku 4.5 | $0.00001 | $0.01421 |
Grade A, and why
fine-tuning-mlp scanned grade A with 1 finding 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 12d 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.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
subprocess.run(cmd, check=True) How it starts
The opening of the file, as written. The whole thing — 954 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Fine-Tuning Machine-Learned Interatomic Potentials
Description
This skill covers fine-tuning pretrained universal machine-learned interatomic potentials on system-specific DFT data: when fine-tuning is the right strategy, how to prepare a small targeted dataset, how to configure MACE's --foundation_model workflow, how to handle reference energies and label offsets, how to avoid catastrophic forgetting, and how to validate and deploy the resulting model. Invoke this skill after universal-potentials benchmarking has confirmed that a pretrained model (MACE-MP-0, CHGNet, SevenNet, ORB) falls short of the accuracy target for a specific chemistry or structural regime, and before committing to full system-specific training from scratch.
Domain Context
Transfer learning for MLPs exploits a pretrained model's learned representation of atomic environments — built from millions of DFT calculations across the periodic table — as a starting point for a new, narrowly defined task. The key physical insight is that local chemical environments for a new target system (e.g., a W-Re alloy, a ZnO surface, an organic solvent at 400 K) are structurally similar to environments the pretrained model has already seen in the training database. Fine-tuning adjusts the model weights to improve accuracy within the target domain while ideally retaining the broad coverage of the pretrained backbone.
Why fine-tuning works. A universal MACE-MP-0 model trained on ~150,000 Materials Project structures has learned to embed atomic environments in a high-dimensional space that already separates coordination geometries, element types, and bonding patterns. The final readout layers (energy and force heads) map these embeddings to energies and forces. For a new target system with similar chemistry, only the readout layers and possibly the last interaction block need updating. This is analogous to transfer learning in computer vision: the early convolutional filters (low-level feature detectors) transfer; the classification head does not.
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.
- 12d ago First seen · 954 lines · 6 tokens per session scan A 061ef74c8986
fine-tuning-mlp is a skill published in the GitHub repository SFETNI/Deep-Matter-Chem-Skills (6 stars, last pushed 1mo ago), licensed MIT. It adds 6 tokens to every session and 14,205 once invoked, about $0.0000 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
primekg
Query the Precision Medicine Knowledge Graph (PrimeKG) for multiscale biological relationships across genes and proteins, drugs, diseases, phenotypes, pathways, biological processes, exposures and anatomy. Use this skill to search entities by name, pull direct neighbours and their evidence types, summarise the local…
fragment-based-count-matrix-generation
Use when you have a backed AnnData object containing processed fragment data (stored in .obsm['fragmentpaired'] or .
methylbase-object-handling
Use when after reading in per-sample methylation call files with methRead() and obtaining methylRawList objects, but before calculating differential methylation or performing annotation.
motif-annotation-correlation-analysis
Use when you have a chromVARDeviations object with multiple annotation sets (such as JASPAR motifs and kmers) and need to determine which annotation pairs are redundant (high correlation) versus synergistic (high synergy z-scores).
motif-database-query-and-matching
Use when you have a set of differentially accessible peaks (output from differential accessibility testing, e.g., tl.
motif-enrichment-statistical-testing
Use when after identifying a set of differentially accessible peaks (via tl.difftest or equivalent), when you need to infer which transcription factors may regulate the observed chromatin state changes.