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 agentmods add skills/learningmatter-mit/atomisticskills/chem-msms-predictnpx skills add learningmatter-mit/AtomisticSkills --skill chem-msms-predictgit clone --depth 1 https://github.com/learningmatter-mit/AtomisticSkillsWhat 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 | $0.00051 | $0.01540 |
| Opus 5 | $0.00026 | $0.00770 |
| Sonnet 5 | $0.00010 | $0.00308 |
| Haiku 4.5 | $0.00005 | $0.00154 |
Grade A, and why
chem-msms-predict 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 3d 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 — 139 lines — stays where its author put it; the contents beside it link to each section on GitHub.
LC-MS/MS Spectrum Prediction
Goal
Predict the LC-MS/MS (tandem mass) spectrum of a molecule given its SMILES string using ICEBERG — a two-stage GNN that first generates a fragmentation DAG (fragment ions) and then predicts their intensities. Output is a predicted spectrum (m/z, intensity) with optional fragment SMILES assignments per peak.
When to Use This Skill
- A SMILES string is known and a predicted LC-MS/MS spectrum (m/z vs intensity) is needed.
- Fragment ion assignments (SMILES per peak) are required.
- No reference spectrum exists, or comparison to a predicted spectrum is desired.
- Companion skill
chem-spectrum-matchercan compare predicted vs experimental spectra.
When NOT to Use This Skill
- Experimental spectrum already available — use it directly; no prediction needed.
- Only compound name known — first resolve to SMILES via
drug-db-pubchem, then call this skill. - GC-MS or other MS types — ICEBERG is trained on LC-MS/MS only; flag a warning before proceeding.
- Organometallics or MW > 1000 — predictions may be unreliable or fail due to unsupported element types.
Prerequisites
1. Download ICEBERG checkpoints
Download from coleygroup/ms-pred releases and place in downloads/:
downloads/
├── iceberg_dag_gen_msg_best.ckpt # generator (stage 1)
└── iceberg_dag_inten_msg_best.ckpt # intensity predictor (stage 2)
Flag error and stop if either checkpoint is missing.
2. Set up the conda environment
bash conda-envs/msms-agent/install.sh
The ms_pred Python package is installed from GitHub automatically by the install script.
Instructions
Step 1 — Run inference and generate spectrum
# Env: ms-gen
python .agents/skills/chem-msms-predict/scripts/predict_msms.py \
--smiles "c1ccccc1C(=O)OCCN" \
--gen_ckpt downloads/iceberg_dag_gen_msg_best.ckpt \
--inten_ckpt downloads/iceberg_dag_inten_msg_best.ckpt \
--collision_energies 20 40 \
--adduct "[M+H]+" \
--instrument "Orbitrap" \
--output_dir results/msms_prediction
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 3d ago First seen · 139 lines · 51 tokens per session scan A 409fda33fcd1
chem-msms-predict is a skill published in the GitHub repository learningmatter-mit/AtomisticSkills (154 stars, last pushed 8d ago), licensed MIT. It adds 51 tokens to every session and 1,540 once invoked, about $0.0003 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
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…
datamol
Pythonic wrapper around RDKit with simplified interface and sensible defaults. Preferred for standard drug discovery including SMILES parsing, standardization, descriptors, fingerprints, clustering, 3D conformers, parallel processing. Returns native rdkit.Chem.Mol objects. For advanced control or custom parameters…
hypogenic
Automated LLM-driven hypothesis generation and testing on tabular datasets. Use when you want to systematically explore hypotheses about patterns in empirical data (e.g., deception detection, content analysis). Combines literature insights with data-driven hypothesis testing. For manual hypothesis formulation use…
binding-affinity
Hybrid ML + physics binding affinity prediction. Empirical scoring, MM/GBSA rescoring, multi-method consensus, and batch virtual screening for protein-ligand complexes.
drug-design
End-to-end drug discovery pipeline orchestration. Deterministic Python script that auto-chains structure prediction, pocket detection, de novo design, docking, scoring, and ADMET filtering into reproducible workflows.
molecular-optimization
Iterative lead optimization with analyze-reason-generate-verify-evaluate loop. Paper-backed (MT-Mol, DrugR, MultiMol).