Claude Skills
The open format is called Agent Skills and works in Claude Code, Codex, Cursor and other agents — most people know it as Claude Skills.
Every Agent Skill we could find on GitHub, deduplicated by content. 79 870 files from 1 769 authors, of which 62 217 are unique — the rest is the same skill repackaged into someone else's repository. For each one: what it weighs in tokens, whether it ships runnable scripts, and which MCP servers it needs.
> Query the ChEMBL database for drug molecules, bioactivity data, and drug targets via the ChEMBL REST API. Use whenever the user asks about drug properties (molecular weight, logP, Lipinski violations), drug-target interactions, bioactivity assay results, or wants to look up any entity by ChEMBL ID or drug/gene name in ChEMBL. Supports single entity or batch queries. No API key required.
> Query DailyMed for FDA drug label / package insert information. Use whenever the user asks about drug labeling, SPL documents, prescribing information, NDC codes, or needs to look up current FDA-approved drug details by name or NDC. Supports single entity or batch queries.
> Query the DDInter drug-drug interaction database. Use whenever the user asks about drug-drug interactions, DDI severity levels, or wants to look up interactions for a drug name or DDInter ID.
> Query the DGIdb (Drug-Gene Interaction Database) for drug-gene interactions, gene druggability categories, and drug target information. Use whenever the user asks about drug targets, druggable genes, gene-drug interactions, or wants to look up any entity (gene name, drug name, druggability category) in DGIdb.
> Query the DILIrank/FDA Liver Toxicity Knowledge Base (LTKB). Use whenever the user asks about drug-induced liver injury (DILI) risk, hepatotoxicity classification, or wants to look up any drug (by name, LTKB ID, or DILIst ID) in the DILIrank or DILIst datasets.
> Query the DRKG (Drug Repurposing Knowledge Graph). Use whenever the user asks about drug–gene, drug–disease, gene–disease, or other biomedical entity relationships in a knowledge-graph context, drug repurposing candidates, COVID-19 drug repurposing, or wants to explore neighbours of any biomedical entity (compound, gene, disease, pathway, side effect, etc.) in DRKG.
> Query a locally downloaded DrugBank database. Use whenever the user asks about drug information, drug targets, drug-drug interactions, drug categories, or wants to look up any entity (DrugBank ID, drug name, CAS number, synonym) in DrugBank.
> Query the DrugCentral drug pharmacology database. Use whenever the user asks about approved drug structures, drug targets, pharmacological actions, or wants to look up any entity (drug name, DrugCentral ID, CAS number, InChIKey) in DrugCentral.
> Query the DrugComb drug combination database for cancer cell-line synergy and sensitivity data. Use whenever the user asks about drug combinations, synergy scores (ZIP/Bliss/Loewe/HSA), combination sensitivity (CSS), or wants to look up how two drugs interact in a specific cancer cell line.
> Query canonical DrugCombDB combination records. Use when the user asks about drug pairs, synergy values, or cell-line-specific combination evidence.
> Query the DrugMechDB drug mechanism-of-action database. Use whenever the user asks about drug mechanisms, drug-to-disease paths, biological targets of a drug, or wants to look up any biomedical entity (drug name, protein, disease, DrugBank ID, MESH ID, UniProt ID, GO term, etc.) in DrugMechDB.
> Query the DrugRepoBank drug repurposing evidence database. Use whenever the user asks about repurposing candidates, drug–disease–target repurposing evidence, or wants to look up any entity (drug name, DrugBank ID, ChEMBL ID, PubChem CID, TTD target ID, UniProt ID, disease name) in DrugRepoBank.
> Query the DrugLib.com Drug Review Dataset (UCI #461). Use whenever the user asks about patient drug reviews, drug effectiveness ratings, side-effect profiles, or condition-specific treatment experiences from DrugLib.com.
> Query the FDA Adverse Event Reporting System (FAERS) via openFDA API. Use whenever the user asks about adverse drug reactions, side effects, drug safety signals, or wants to look up reported adverse events for one or more drug names.
> Query or inspect the FDA Orange Book - FDA-Approved Drug Products Listing resource for drug-centric tasks with emphasis on drug knowledgebase Use whenever Codex needs the calling pattern, downloadable entrypoint, or example query flow from this skill example script.
> Query the Gene-Drug Knowledge Database (GDKD) for variant-specific gene–drug associations in oncology. Use when the user asks about cancer genomic biomarkers, drug sensitivity/resistance by gene or variant, targetable mutations, or clinical evidence for cancer therapeutics.
> Query the IUPHAR/BPS Guide to Pharmacology REST API for drug targets, ligands (drugs/compounds), and their interactions. Use whenever the user asks about pharmacological targets, receptor–ligand relationships, drug mechanisms of action, or wants to look up any drug or target name in IUPHAR. Supports single entity or batch queries. No API key required.
> Query the MecDDI mechanism-based drug-drug interaction database. Use whenever the user asks about drug-drug interactions, DDI mechanisms (PK/PD), enzyme or transporter-mediated interactions, or wants to look up interacting drug pairs by drug name or MecDDI drug ID. Trigger on keywords like DDI, drug interaction, MecDDI, mechanism-based interaction, pharmacokinetic interaction, pharmacodynamic interaction, or any query involving two drugs that may interact.
> Query MedlinePlus for consumer-oriented drug and health-topic information. Accepts drug names, RxCUI codes, NDC codes, or ICD-10-CM diagnosis codes. MedlinePlus Connect (code-based lookup).
> Query the NCI CCDI Molecular Targets Platform (pediatric oncology) for targets (genes), diseases, drugs, and target-disease associations via its public GraphQL API. Auto-detects entity type from input string.
> Query the NCI-60 Molecular Target (Protein) database from the Developmental Therapeutics Program. Use when the user asks about protein expression of drug targets across the NCI-60 cancer cell line panel, or wants to look up a gene, cell line, or cancer panel in the NCI DTP molecular target dataset.
> Query NDF-RT (National Drug File Reference Terminology) via the NCI EVS REST API. Use when looking up drug mechanisms of action, physiological effects, pharmacologic classes, chemical structures, or drug–disease relationships (may_treat / may_prevent) in NDF-RT. Accepts drug names or NDF-RT codes.
> Query the nSIDES drug side effect databases (OnSIDES, OffSIDES, KidSIDES). Use whenever the user asks about drug adverse reactions, side effects, off-label safety signals, or pediatric drug safety for a given drug name.
> Query the Open Targets Platform for drug-target-disease associations. Use whenever the user asks about drug targets, gene-disease associations, drug indications, clinical trial phases, or wants to look up any entity (Ensembl gene ID, ChEMBL drug ID, or free-text gene/drug name) in Open Targets. Also trigger when the user mentions Open Targets, ENSG IDs, CHEMBL IDs, or asks about target prioritization for diseases.
> Query FDA drug labeling data via openFDA. Use whenever the user asks about drug prescribing information — indications, warnings, dosage, adverse reactions, contraindications, or administration routes. Supports single or batch lookup by brand/generic name or by indication/condition.
> Query the OREGANO knowledge graph for computational drug repurposing. Use whenever the user asks about drug–target–disease–gene–pathway relationships, compound cross-references, drug repurposing hypotheses, or wants to explore neighbors of any biomedical entity in a knowledge graph that includes natural compounds.
> Query ClinPGx (PharmGKB) and CPIC for pharmacogenomics data. Use whenever the user asks about gene-drug interactions, pharmacogenomics clinical annotations, drug-metabolizing enzymes, CPIC guidelines, or variant-level PGx evidence for any gene symbol, drug name, rsID, or ClinPGx accession.
> Query the PharmKG knowledge graph (180k entities, 39 relation types, >1M triples). Use whenever the user asks about biomedical relationships among genes, drugs/chemicals, and diseases — e.g. drug–gene interactions, drug–disease associations, gene–disease links, or drug–drug relationships derived from literature and curated databases.
> Query the PHEE pharmacovigilance event extraction dataset. Use whenever the user asks about annotated adverse drug events, pharmacovigilance case reports, drug–effect associations from medical literature, or wants to find PHEE examples mentioning a drug name, adverse effect, or condition.
> Query the PsyTAR psychiatric adverse-reaction corpus. Use when the user asks about patient-reported ADRs, withdrawal symptoms, drug indications, or effectiveness for Zoloft, Lexapro, Cymbalta, or Effexor XR. Accepts drug names (brand or generic), symptom terms, or UMLS CUIs.
> Query the RepoDB drug repurposing database. Use whenever the user asks about drug-disease associations, drug repurposing candidates, or wants to look up any entity (drug name, indication, DrugBank ID, UMLS CUI, NCT ID) in RepoDB.
> Query the RepurposeDrugs single-agent drug repurposing database. Use whenever the user asks about drug-disease repurposing associations, clinical trial phases for repurposed drugs, or wants to look up any entity (drug name, disease name, NCT ID) in RepurposeDrugs.
> Query the Broad Institute Drug Repurposing Hub (~6,800 compounds). Look up drugs by name, gene target, MOA, disease area, Broad ID, or InChIKey. Returns clinical phase, mechanism of action, targets, disease area, indication, and chemical identifiers.
> Query the RxNorm drug naming and normalization API. Use whenever the user asks to look up an RxCUI, normalize a drug name, find drug interactions, retrieve brand/trade names, or resolve any clinical drug name via RxNorm. Supports single drug or batch queries. Trigger on mentions of RxNorm, RxCUI, drug normalization, drug interaction lookup, or brand-name resolution.
> Query the STITCH chemical-protein interaction database. Use whenever the user asks about chemical-protein interactions, drug-target binding, compound action modes, or wants to look up any entity (chemical name, STITCH CID, STRING protein ID) in STITCH.
> Query TAC 2017 ADR annotated drug labels for adverse drug reactions. Use whenever the user asks about ADRs extracted from FDA drug labels, MedDRA-normalized adverse reactions, or wants to look up a drug name, ADR string, or MedDRA code in the TAC 2017 ADR corpus.
> Query canonical TarKG drug-target triplets. Use when the user asks about drug-target interactions, relation labels, disease/pathway context, or quick lookups for drugs/targets in TarKG.
> Query the Therapeutic Target Database (TTD) for drug-target-disease interaction data. Use this skill when the user asks about therapeutic targets, drugs, diseases, or their relationships — including target-drug mappings, clinical status of drugs, disease indications, UniProt/gene associations, and pathway annotations. Triggers on queries like "what drugs target EGFR", "which diseases is Imatinib used for", "find targets for lung cancer", or any lookup involving TTD IDs, gene symbols, drug names, or disease names.
> Query the UniTox drug toxicity database. Use whenever the user asks about organ-system toxicity ratings for a drug, multi-organ toxicity profiles, or wants to look up any entity (drug name, SMILES, SPL_ID) in UniTox.
> Query the WebMD Drug Reviews dataset (~362 k patient reviews, 2007–2020). Use whenever the user asks about patient-reported drug effectiveness, ease of use, satisfaction ratings, side effects, or reviews for a specific drug or medical condition.
> Query the WHO Model List of Essential Medicines (23rd list, 2023). Use whenever the user asks about essential medicines, WHO-recommended drugs, dosage forms, therapeutic sections, or AWaRe antibiotic classification.
Use this skill for creating or refining an academic slide deck and the talk built around it: structuring a conference talk, thesis defense, lab meeting, or paper-to-slides deck; deciding the narrative arc and slide breakdown; improving slide design and visual hierarchy; planning rehearsal, timing, Q&A, and backup slides; or generating the .pptx. Reach for it when the user is shaping the presentation itself. Do not use for writing the paper, producing standalone speaker notes/scripts/transcripts, making posters, creating isolated figures/charts outside a slide deck, or building non-academic presentations.
Manages persistent research memory across ideation and experimentation cycles. Maintains two stores: Ideation Memory M_I (feasible/unsuccessful directions) and Experimentation Memory M_E (reusable strategies for data processing, model training, architecture, debugging). Three evolution mechanisms: IDE (after research-ideation), IVE (after experiment failure — classifies failures as implementation vs fundamental), ESE (after experiment success — extracts reusable strategies). Use when: updating memory after completing research-ideation cycles or experiment pipelines, classifying why a method failed (implementation vs fundamental failure), starting a new research cycle needing prior knowledge, user mentions 'update memory', 'classify failure', 'what worked before', 'research history', 'evolution'. Do NOT use for running experiments (use experiment-pipeline), debugging experiment code (use experiment-craft), or generating ideas (use research-ideation).
Use this skill whenever the user submits a non-trivial mathematical claim that needs a rigorous proof or audit. Trigger on IMO/Putnam/USAMO/Olympiad-style problems, ML/AI theoretical statements, research conjectures, suspected-false claims, multi-step proofs the user already failed on, proof drafts with possible hidden assumptions, or any request containing 'prove rigorously', 'verify this', 'is this true', 'find the gap', 'audit my proof', 'find a counterexample', or 'use EvoMath' that targets a mathematical claim. Activate also when the problem requires more than three reasoning steps. Do NOT use for single-step calculations, definition lookups, textbook exercises with a known recipe, code analysis tasks, literature survey questions, pure symbolic manipulation, or non-mathematical applications of those trigger phrases (e.g., 'is it true that GPT-4 can solve math?', 'verify this LaTeX syntax'); hand those back instead.
Use this skill when the user wants to debug, diagnose, or systematically iterate on an experiment that already exists, or when they need a structured experiment log for tracking runs, hypotheses, failures, results, and next steps during active research. Apply it to underperforming methods, training that will not converge, regressions after a change, inconsistent results across datasets, aimless experimentation without progress, and questions like 'why doesn't this work?', 'no progress after many attempts', or 'how should I investigate this failure?'. Also use it for setting up practical experiment logging/record-keeping that supports debugging and iteration. Do not use it for designing a brand-new experiment pipeline or full experiment program (use experiment-pipeline), generating research ideas, fixing isolated coding/syntax errors, or writing retrospective summaries into research memory/notes/knowledge bases.
Iterative code refinement through plan → code → evaluate → refine cycles. Runs lint checks (ruff), tests (pytest), and structured self-evaluation each cycle, then diagnoses failures and refines. Decomposes complex tasks into sequential phases, iterates up to 3 times per phase (10 total). Use when: the main agent delegates a code task with 'MODE: MORE_EFFORT', the user selects 'More Effort' code generation mode, or the task explicitly requests iterative refinement for higher code quality. Do NOT use for single-pass code generation (Lite mode), experiment pipeline orchestration (use experiment-pipeline), or diagnosing a specific experiment failure (use experiment-craft).
Guides structured 4-stage experiment execution with attempt budgets and gate conditions: Stage 1 initial implementation (reproduce baseline), Stage 2 hyperparameter tuning, Stage 3 proposed method validation, Stage 4 ablation study. Integrates with evo-memory (load prior strategies, trigger IVE/ESE) and experiment-craft (5-step diagnostic on failure). Use when: user has a planned experiment, needs to reproduce baselines, organize experiment workflow, or systematically validate a method. Do NOT use for debugging a specific experiment failure (use experiment-craft) or designing which experiments to run (use paper-planning).
Generate professional presentation slides and high-quality illustrations using Gemini image generation API (Nano Banana 2), with interactive browser-based review and iterative editing. Full workflow: content planning conversation → slides_plan.json → batch image generation → review with feedback → targeted slide editing → PPTX packaging. Use when: user wants to create a presentation, make slides, generate a PPT/PPTX, prepare a talk deck, design visual slide content, or generate high-quality figures/illustrations for papers and documents. Do NOT use for: writing academic papers (use paper-writing) or planning academic conference talk narrative structure (use academic-slides).
Find, read, download, and locally cache academic papers. Disambiguate ambiguous queries, discover via keyword search / citation traversal / recommendations / arXiv monitoring / trending / GitHub search, evaluate (TLDR, citations, code, SOTA), read using a 3-level strategy, and save PDFs to a local library for offline reuse. Use when finding a specific paper, listing papers on a topic, tracking recent advances, finding a baseline with code, reading or downloading a paper by URL, searching the local PDF library, or collecting a corpus for survey/ideation. Trigger phrases include: find/search papers, related work, citation analysis, latest research, download paper, save paper, my local library. Do NOT use for generating survey reports (use research-survey), generating research ideas (use research-ideation), writing a Related Work section (use paper-writing), comparing/ranking ideas (use research-ideation), or planning paper structure (use paper-planning).
Guides pre-writing planning for academic papers with 4 structured steps: story design (task-challenge-insight-contribution-advantage), experiment planning (comparisons + ablations), figure design (pipeline + teaser), and 4-week timeline management. Includes counterintuitive planning tactics (write a mock rejection letter to identify weaknesses before writing, narrow before broad claims, design ablations first). Use when: user wants to plan a paper before writing, design story/contributions, plan experiments, create figure sketches, set a writing timeline, or write a pre-emptive rejection letter for planning purposes. Do NOT use for actual writing (use paper-writing), running experiments (use experiment-pipeline), self-reviewing a finished draft (use paper-review), or finding research problems (use research-ideation).
Claude Skills — questions
Answers built from the skills we actually parsed.