Skills.

Give your AI a better way to work.

A skill is a set of written instructions that teaches an AI how to do one job the way it should be done: review a pull request, plan a migration, write the release notes.

Install one here and it travels with your account into Claude, Claude Code, Cursor and every other client you sign in with.

Filter by category

55,435 results · page 922 of 1,848

  • mat-random-structure-searchSkillSearch

    Lets your agent generate and relax random crystal structures to find low-energy candidates for a chemical composition.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • mat-reaction-networkSkillDev tools

    Lets your agent predict the best solid-state synthesis pathways for inorganic materials and list the reactions.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • mat-sample-pes-by-mdSkillDev tools

    Lets your agent sample off-equilibrium potential energy surfaces to benchmark and fine-tune machine learning potentials.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • mat-solid-free-energySkillDev tools

    Lets your agent calculate the Helmholtz free energy of a periodic solid structure using MLIP models.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • mat-stabilitySkillDev tools

    Lets your agent calculate a material's thermodynamic stability and its energy above the convex hull at 0K.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • mat-structure-noveltySkillDev tools

    Lets your agent check whether a material structure matches known ones or compare two structures.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • mat-surface-adsorptionSkillDev tools

    Lets your agent calculate how strongly molecules stick to material surfaces using machine learning models.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • mat-surface-energySkillDev tools

    Lets your agent calculate surface energy of crystal planes and build equilibrium Wulff shapes.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • mat-synthesis-extractionSkillFiles & storage

    Lets your agent extract structured synthesis procedures from PDF papers into JSON records.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • mat-synthesis-recommendationSkillDatabases & data

    Lets your agent search and rank materials synthesis recipes with precursors, procedures, and journal references.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • mat-xrd-calculatorSkillDev tools

    Lets your agent calculate a material's X-ray diffraction spectrum using pymatgen.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • mat-xrd-digitizerSkillFiles & storage

    Lets your agent turn an image of an XRD plot into a numeric .xy data file by extracting its peaks.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • mat-xrd-phase-analysisSkillSearch

    Lets your agent identify material phases from experimental XRD data using tree search.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • mat-xrd-refinementSkillDev tools

    Lets your agent refine crystal structures by fitting computed patterns to experimental XRD data using Rietveld refinement.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • ml-autoresearchSkillAI & models

    Lets your agent run machine learning training experiments that analyze results like gradients and errors before deciding the next change.

    Ready to connect★ 172

    github.com/gaasher/agent-loop-skills172 stars

    View details
  • ml-bayesian-optimizationSkillDev tools

    Lets your agent find the best settings for costly experiments or simulations by learning from each past result.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • ml-cluster-expansionSkillDev tools

    Lets your agent train a cluster expansion model for simulating disordered materials with Monte Carlo.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • ml-committee-uncertaintySkillAI & models

    Lets your agent measure how uncertain an ML potential's predictions are and flag structures needing verification.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • ml-fairchem-finetuneSkillDatabases & data

    Lets your agent fine-tune Fairchem machine learning models that predict atomic interactions on your own datasets.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • ml-foundation-potentialsSkillAI & models

    Helps your agent pick the right machine-learning atom model for a materials simulation.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • ml-generative-aditSkillAI & models

    Lets your agent generate new crystal structures and molecules using an all-atom diffusion model.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • ml-generative-diffcspSkillDev tools

    Lets your agent generate crystal structures for a chosen composition using the DiffCSP++ model.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • ml-generative-mattergenSkillAI & models

    Lets your agent generate new inorganic material structures using a diffusion-based generative model.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • ml-mace-finetuneSkillDatabases & data

    Lets your agent fine-tune MACE machine learning models that simulate atomic interactions on custom datasets.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • ml-matgl-finetuneSkillDatabases & data

    Lets your agent fine-tune MatGL machine learning models for simulating atomic interactions on custom datasets.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • ml-mlip-automlSkillSearch

    Lets your agent automatically tune hyperparameters for molecular machine learning models like MACE, MatGL, and FairChem.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • ml-mlip-benchmarkSkillDatabases & data

    Lets your agent test machine learning force field accuracy by computing energy and force errors and plotting results.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • ml-mlip-nvalchemiSkillAI & models

    Lets your agent run fast batched simulations of materials, including relaxations and molecular dynamics, on GPU.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • ml-mlip-speedSkillAI & models

    Lets your agent look up benchmark results comparing how fast different machine learning interatomic potentials run.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details
  • ml-property-predict-scdSkillAI & models

    Lets your agent train a model to predict properties of molecules or periodic materials.

    Ready to connect★ 172

    github.com/learningmatter-mit/atomisticskills158 stars

    View details

What is a skill?

A skill is plain text, usually a SKILL.md file and the scripts it refers to, written for an AI rather than for a person. It carries the steps, the house rules and the examples a good answer needs, so you stop pasting the same briefing into every new chat.

55,004 of the 55,435 skills listed here can be served through ahel today, and they come from public repositories. Each one has its own page with the instructions themselves on it, so you can read what a skill will tell your AI to do before you install it.

Install one and every AI you use gets it

Installing a skill adds it to your gateway and turns it on in the same step. Claude Code surfaces it as a slash command; any client can read the full instructions with the skill_read tool.

Nothing is copied into a project folder. The instructions are served from your account, so the same skill is there in every AI you connect, and turning it off removes it from all of them at once.

See how to connect your AI