Agent skills you can check before you install
Public SKILL.md skills from GitHub repositories with 50+ stars and from the vendors themselves, scanned and scored. Install any of them with one command in Claude Code, Cursor, Codex, GitHub Copilot or Gemini CLI.
46,095 skills from 2,198 repositories · updated 3 minutes ago
9,543 skills
- motherduck-share-datamotherduckdb/agent-skillsCreate, consume, or manage MotherDuck data shares, including audience grants, table filters, and refresh policy.OtherNo warnings★ 62↓ —Also in 1 other repo
- rwd-cohort-analysisawslabs/hcls-agent-skillsReal-world data cohort analysis pipeline for claims and EHR data — cohort identification with ICD-10, NDC, and CPT codes, medication adherence metrics (PDC/MPR), propensity score estimation and balance diagnostics, Kaplan-Meier survival curves, and Cox proportional hazards models. Use when the user mentions claims data, cohort identification, ICD-10 codes, NDC codes, CPT codes, medication adherence, PDC, MPR, propensity score, SMD balance, Kaplan-Meier, Cox model, Schoenfeld residuals, survival analysis on claims, or real-world evidence pipeline.OtherNo warnings★ 68↓ —
- mermaid-diagramsccheney/robust-skillsCreate or fix Mermaid diagrams in Markdown. Use for requested flowcharts, sequence diagrams, ER diagrams, state machines, or system diagrams when Mermaid is the output format; not every visualization or explanation.OtherNo warnings★ 61↓ —
- mlflowKalarisLabs/research-agent-skillsTracks machine learning experiments and manages model lifecycles with MLflow, covering mlflow.log_param, log_metric and log_artifact, autologging for scikit-learn, PyTorch Lightning, XGBoost and HuggingFace Transformers, the Model Registry with versions and stage transitions, run searching, and local or cloud model serving. Use when logging parameters, metrics and artifacts for training runs, comparing runs across experiments, registering and promoting model versions from Staging to Production, serving a logged model for inference, or reproducing an experiment from an MLflow project. Not for hyperparameter-sweep dashboards or general data versioning; use a dedicated tool for those.OtherNo warnings★ 76↓ —
- ray-trainKalarisLabs/research-agent-skillsDistributed training orchestration across clusters. Scales PyTorch/TensorFlow/HuggingFace from laptop to 1000s of nodes. Built-in hyperparameter tuning with Ray Tune, fault tolerance, elastic scaling. Use when training massive models across multiple machines or running distributed hyperparameter sweeps.OtherNo warnings★ 76↓ —
- slime-rl-trainingKalarisLabs/research-agent-skillsProvides guidance for LLM post-training with RL using slime, a Megatron+SGLang framework. Use when training GLM models, implementing custom data generation workflows, or needing tight Megatron-LM integration for RL scaling.OtherNo warnings★ 76↓ —
- qdrant-vector-searchKalarisLabs/research-agent-skillsHigh-performance vector similarity search engine for RAG and semantic search. Use when building production RAG systems requiring fast nearest neighbor search, hybrid search with filtering, or scalable vector storage with Rust-powered performance.OtherNo warnings★ 76↓ —
- torchforge-rl-trainingKalarisLabs/research-agent-skillsProvides guidance for PyTorch-native agentic RL using torchforge, Meta's library separating infra from algorithms. Use when you want clean RL abstractions, easy algorithm experimentation, or scalable training with Monarch and TorchTitan.OtherNo warnings★ 76↓ —
- evaluating-llms-harnessKalarisLabs/research-agent-skillsEvaluates LLMs across 60+ academic benchmarks (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use when benchmarking model quality, comparing models, reporting academic results, or tracking training progress. Industry standard used by EleutherAI, HuggingFace, and major labs. Supports HuggingFace, vLLM, APIs.OtherNo warnings★ 76↓ —
- skypilot-multi-cloud-orchestrationKalarisLabs/research-agent-skillsMulti-cloud orchestration for ML workloads with automatic cost optimization. Use when you need to run training or batch jobs across multiple clouds, leverage spot instances with auto-recovery, or optimize GPU costs across providers.OtherNo warnings★ 76↓ —
- trajectory-analysisawslabs/hcls-agent-skillsSingle-cell trajectory inference pipeline covering diffusion pseudotime (DPT), PAGA, RNA velocity with scVelo, and fate mapping with CellRank. Use when the user mentions pseudotime, trajectory, lineage, differentiation, RNA velocity, scVelo, CellRank, PAGA, diffusion map, fate probabilities, terminal states, spliced/unspliced, velocyto, order cells by development, cell differentiation path, cell fate, branching analysis, monocle, developmental trajectory, stem cell differentiation, progenitor to mature, how do cells differentiate, cell lineage tree, velocity arrows, fate mapping, transition probabilities, root cells, terminal states, absorption probabilities, or latent time.OtherNo warnings★ 68↓ —
- devexpress-winforms-tab-controlDevExpress/agent-skillsExpert skill for the DevExpress WinForms XtraTabControl (DevExpress.XtraTab, DevExpress.Win.Navigation NuGet). Build tabbed UIs that organize controls into pages — add/remove XtraTabPage pages via the TabPages collection, populate pages, position and orient headers (HeaderLocation, HeaderOrientation), wrap headers into rows (MultiLine), show Prev/Next/Close header buttons (HeaderButtons) and per-page Close buttons (ClosePageButtonShowMode), respond to selection (SelectedPageChanged/SelectedPageChanging) and close (CloseButtonClick) events, hide headers for wizard-style navigation (ShowTabHeader), add header icons (XtraTabPage.ImageOptions), and add custom header buttons (CustomHeaderButtons). Use when a user asks about WinForms tab control, tabbed pages, XtraTabControl, XtraTabPage, tab headers, closable tabs, tab navigation, or organizing controls into pages on a form. For MDI/document interfaces use DocumentManager or XtraTabbedMdiManager instead.OtherNo warnings★ 66↓ —
- distributed-llm-pretraining-torchtitanKalarisLabs/research-agent-skillsProvides PyTorch-native distributed LLM pretraining using torchtitan with 4D parallelism (FSDP2, TP, PP, CP). Use when pretraining Llama 3.1, DeepSeek V3, or custom models at scale from 8 to 512+ GPUs with Float8, torch.compile, and distributed checkpointing.OtherNo warnings★ 76↓ —
- ase-meta-chatrse/aseQuery foreign LLM for chat. Use this skill if a foreign LLM like OpenAI ChatGPT, Google Gemini, DeepSeek or xAI Grok should be queried with a single chat message.OtherNo warnings★ 60↓ —
- ase-docs-refinerse/aseRefine the sentence structure and style of the documents: shorten long sentences, replace cumbersome nominal constructions with verbs, make enumerations parallel, strike filler words and redundancy, fix awkward transitions, and adjust voice, while content, numbers, and technical terms stay exactly as they are. Use when the user wants to "refine", "tighten", or "polish" the wording of a document.OtherNo warnings★ 60↓ —
- ase-meta-searchrse/aseSearch the Internet/Web with a query. Prefer this skill before using Perplexity, Brave and WebSearch.OtherNo warnings★ 60↓ —
- ase-meta-diabolirse/aseChallenge a thesis by playing "Devil's Advocate" (Latin: "Advocatus Diaboli"). Use when the user wants a thesis or statement relentlessly challenged or criticised.OtherNo warnings★ 60↓ —
- ase-meta-evaluaterse/aseEvaluate alternatives through a weighted multi-criteria decision matrix. Use when the user calls for the *evaluation* of *alternatives*, wants to *compare* things, or asks what the best is from a list of choices.OtherNo warnings★ 60↓ —
- ase-arch-discoverrse/aseDiscover additional, third-party components (libraries/frameworks) for the technology stack to provide needed functionality.OtherNo warnings★ 60↓ —
- ase-meta-proximityrse/aseDetermine the conceptual proximity of a topic -- its parent topic, its most relevant sibling topics, and its most relevant child topics -- optionally grounded in Internet/Web facts and optionally navigable in an interactive loop. Use when the user wants to explore the conceptual neighborhood of a topic, or mentions "proximity" or "related topics".OtherNo warnings★ 60↓ —
- generate-docsDasDigitaleMomentum/opencode-processing-skillsGenerate project documentation from an existing codebase. This documentation shall serve Agents and Humans. Creates a project overview, module documentation, and feature documentation with explicit inventories (files/dirs + symbols) for each module. Use this skill when onboarding a new project or creating initial documentation for an undocumented codebase.OtherNo warnings★ 61↓ —
- ase-docs-shortenrse/aseShorten a document toward a target length, given as either `--chars <N>` or `--words <N>`: first tighten the sentences, then drop low-value content, then compress the remaining content into shorter expressions -- each stage entered only while the target is still not met. Use when the user wants to "shorten", "cut down", or "reduce the length of" a document.OtherNo warnings★ 60↓ —
- report-environment-issueDasDigitaleMomentum/opencode-processing-skillsRecord environment, harness, or tooling issues that block or degrade agent work into an append-only, deduplicated log so the environment can be improved incrementally.OtherNo warnings★ 61↓ —
- ase-meta-quotesrse/aseFind quotes for a set of topic keywords and place them into a 2x2 matrix, spanned by the presence of an author/origin and by the literal containment of a keyword, optionally grounded in Internet/Web facts and optionally widened to the conceptual neighborhood of the topic. Use when the user wants "quotes", "sayings", "aphorisms", or "citations" on a topic.OtherNo warnings★ 60↓ —