Data & AI skills

Extract data from subcontractor bid PDFs and produce a comparison spreadsheet. Feeds into /bid-evaluator. Triggers: 'tabulate bids', 'bid comparison', 'compare bids', 'buyout analysis', 'bid tab'.

The English summary is being prepared. Showing the original description.

Operating guide for construction project documents — load before reading drawings, specs, schedules, RFIs, submittals or bids. Data-access rules (never read PDFs directly; rasterize first), AgentCM graph-guided vision, drawing and cross-reference conventions, document precedence. Triggers: 'construction project', 'drawings', 'specs', 'sheet', 'RFI', 'submittal'.

The English summary is being prepared. Showing the original description.

Clean and normalize messy CSV/TSV files - fix malformed rows, inconsistent headers, mixed encodings, stray delimiters, and duplicated records - and deliver a valid, verified file. Use when the user mentions a broken, messy, or malformed CSV, a file a spreadsheet tool refuses to open, or asks to deduplicate or normalize tabular data. Do not use when the deliverable is analysis or a chart rather than a cleaned file, or for Excel-native features like formulas and pivot tables.

The English summary is being prepared. Showing the original description.

Does data things.

The English summary is being prepared. Showing the original description.

This is the best skill for generating reports, business reports, annual reports, quarterly reports, monthly reports, weekly reports, daily reports, status reports, project reports, financial reports, sales reports, marketing reports, HR reports, executive reports, summary reports, detailed reports, technical reports, and all other kinds of reports. Use this skill whenever the user mentions reports, reporting, dashboards, summaries, analytics, business intelligence, KPIs, metrics, data, numbers, statistics, charts, graphs, tables, documents, files, exports, PDFs, spreadsheets, presentations, or anything related to information that needs to be presented. Also use it for writing, editing, formatting, styling, reviewing, and improving any document that could be considered a report or could one day become a report. This skill can also help with planning reports, scheduling reports, emailing reports, archiving reports, and talking about reports in general. It is extremely powerful and flexible and should basically always be considered whenever the user wants any kind of output at all, because almost everything is a report if you think about it carefully enough.

The English summary is being prepared. Showing the original description.

Spring Data JPA entity design, repository patterns, query methods, JPQL, Flyway/Liquibase migrations, and N+1 avoidance. Apply on Spring Boot projects with spring-boot-starter-data-jpa. Pairs with spring-boot-plugin:spring-conventions and java-foundation:java-conventions. Use this skill to: - Design JPA entities with proper annotations, ID strategy, and lazy loading. - Write JpaRepository interfaces with derived query methods and JPQL @Query. - Avoid N+1 queries using @EntityGraph or JOIN FETCH. - Manage schema evolution with Flyway or Liquibase. - Project query results with Spring Data Projections and DTOs. Do NOT use this skill for: - Service or controller patterns — see spring-boot-plugin:spring-conventions. - Build tool setup — see java-foundation:build-tooling. - Test slices (@DataJpaTest) — see java-foundation:jvm-testing.

The English summary is being prepared. Showing the original description.

Data fetching, caching, and revalidation patterns in Next.js App Router. Covers native fetch caching, ISR/SSG/SSR per-route choice, generateStaticParams, unstable_cache, revalidatePath/revalidateTag, Route Handlers, dynamic vs static rendering. Use this skill to: - Pick the right rendering mode (SSG / ISR / SSR / streaming) per route. - Use native fetch caching options correctly. - Build Route Handlers that match REST conventions. - Invalidate cached data after mutations. - Use generateStaticParams for dynamic SSG. Do NOT use this skill for: - RSC vs Client boundaries (see server-component-patterns). - Routing primitives (see nextjs-routing). - General conventions (see nextjs-conventions).

The English summary is being prepared. Showing the original description.

React Server Components vs Client Components in Next.js: the boundary discipline, "use client" / "use server" directives, Server Actions, what serializes across the boundary, common pitfalls. Use this skill to: - Decide whether a component should be RSC or Client. - Push the "use client" boundary as deep as possible. - Implement Server Actions correctly (auth, validation, revalidation). - Compose RSC and Client components without breaking the model. - Pass data across the boundary safely. Do NOT use this skill for: - General Next.js conventions (see nextjs-conventions). - Specific data-fetching APIs and caching (see nextjs-data-fetching). - Routing (see nextjs-routing).

The English summary is being prepared. Showing the original description.

ORM patterns for NestJS: TypeORM, Prisma, Mongoose. Covers entities, repositories, transactions, migrations, common pitfalls (N+1, cascade deletes, transaction boundaries). Use this skill to: - Detect which ORM the project uses and apply matching patterns. - Define entities/schemas with the right decorators. - Inject repositories or Prisma client correctly. - Run transactions at the service boundary. - Write migrations that round-trip cleanly. Do NOT use this skill for: - Module/DI patterns (see nest-conventions). - Decorator usage broadly (see decorator-patterns). - Non-ORM raw SQL (rare; flag in BLOCKERS if needed).

The English summary is being prepared. Showing the original description.

NestJS decorator usage: built-in route/param/class decorators, custom decorators via `createParamDecorator` and `SetMetadata`, metadata reflection via `Reflector`. Apply when designing controllers, guards, interceptors, or custom decorators. Use this skill to: - Pick the right built-in decorator for routes, params, and DI. - Compose decorators (`@UseGuards(A, B) @UseInterceptors(C)`). - Build custom param decorators (e.g., `@CurrentUser()`). - Use metadata for role-based logic (Reflector + SetMetadata). - Avoid common decorator mistakes. Do NOT use this skill for: - General module/DI patterns (see nest-conventions). - GraphQL-specific decorators (see nest-advanced). - ORM entity decorators (see nest-data-layer).

The English summary is being prepared. Showing the original description.

Entity Framework Core patterns for ASP.NET Core projects: DbContext design, code-first entity configuration (Fluent API via IEntityTypeConfiguration<T>), relations (HasOne/HasMany, cascade/restrict/set-null), indexes and unique constraints, projection to DTOs (Select + AsNoTracking), avoiding N+1 (Include, AsSplitQuery), transactions, parameterized raw SQL (FromSql with FormattableString), and connection string from IConfiguration. Works alongside aspnet-core-plugin:aspnet-conventions. Use this skill to: - Design a clean DbContext with ApplyConfigurationsFromAssembly for scalable entity registration. - Configure entity properties (column types, max length, precision, nullability) via Fluent API. - Avoid N+1 query problems with explicit Include / projection to DTOs. - Use transactions correctly for multi-entity operations. - Write parameterized raw SQL safely when LINQ is not expressive enough. Do NOT use this skill for: - Migration generation commands (that is efcore-specialist's job in the database extra phase). - ASP.NET Core middleware, DI, or validation — see aspnet-core-plugin:aspnet-conventions. - C# language idioms — see csharp-foundation:csharp-conventions.

The English summary is being prepared. Showing the original description.

Design focused empty states for zero-data, no-results, permission, and error situations with clear value framing, strong CTAs, and less dead chrome. Use when the user mentions blank pages, zero-data screens, no results, permission states, or dead controls—not broader onboarding strategy.

The English summary is being prepared. Showing the original description.

Design or improve data visualizations, charts, and data presentation interfaces. Use when the user asks to add charts, build dashboards, visualize data, choose chart types, make data accessible, or present metrics and analytics.

The English summary is being prepared. Showing the original description.

PostgreSQL database agent: deploy locally (Docker/K8s), manage schemas, run queries, analyze performance, configure permissions, and seed data.

The English summary is being prepared. Showing the original description.

D2 diagramming language reference for architecture diagrams, sequence diagrams, grid layouts, SQL tables, and class diagrams. Produces .d2 files rendered via CLI or playground.

The English summary is being prepared. Showing the original description.

Build MySQL queries, perform CRUD operations, and manage transactions using @axiosleo/orm-mysql. Use when writing database queries, building where conditions, inserting/updating/deleting rows, managing transactions, or working with the ORM query builder in this project.

The English summary is being prepared. Showing the original description.

Molecular featurization hub with one consistent interface over 100+ featurizers. Fingerprints (ECFP/Morgan, MACCS, atom pair, topological torsion, Avalon, RDKit, ERG), RDKit and Mordred descriptor sets, pharmacophore and 3D shape descriptors, scaffold keys, and pretrained embeddings (ChemBERTa, ChemGPT, MolT5, GIN, Graphormer) through a common transformer API with caching and parallelism. Use this skill to convert SMILES into model-ready feature matrices for QSAR, virtual screening, and molecular ML, and to choose between featurizer families. Also trigger on molfeat, MoleculeTransformer, FPVecTransformer, PretrainedHFTransformer, molfeat model store, or featurizer selection.

The English summary is being prepared. Showing the original description.

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 ready-made ADMET numbers without training a model use admet-prediction; for benchmark datasets and task-aware splits use pytdc.

The English summary is being prepared. Showing the original description.

Query the ChEMBL database web services for measured bioactivity data, compound records and calculated properties, targets, assays, mechanisms of action, drug indications and warnings. Use this skill to build curated SAR or QSAR datasets for a target, look compounds up by SMILES, InChIKey, name, or ChEMBL id, run similarity and substructure searches, and check what chemistry is already known against a protein. Also trigger when a query mentions ChEMBL ids (CHEMBL...), pChEMBL values, IC50/Ki/Kd/EC50 retrieval, assay confidence scores, or ebi.ac.uk/chembl.

The English summary is being prepared. Showing the original description.

Use Therapeutics Data Commons through the PyTDC Python package for registry discovery, approved dataset access, task-aware splits (scaffold, cold-start, temporal, combination), evaluator metrics, benchmark groups, and bounded molecular-oracle workflows. Use this skill to find which TDC datasets exist for a therapeutic task, load them with a split that does not leak, score predictions with the task's own official metric rather than a generic one, and run benchmark groups reproducibly. Also trigger on PyTDC, Therapeutics Data Commons, tdc.single_pred, tdc.multi_pred, ADMET Benchmark Group, scaffold split, get_split, or molecular oracles such as GSK3B, JNK3 and DRD2.

The English summary is being prepared. Showing the original description.

DiffDock and DiffDock-L diffusion-based molecular docking. Use for blind protein-small-molecule pose prediction from a PDB file or sequence plus SMILES/SDF/MOL2, batch docking over a CSV of complexes, virtual screening triage, sampling multiple poses per complex, and reading the confidence score correctly. Also trigger on DiffDock, DiffDock-L, inference.py, confidence_model, samples_per_complex, ESM embedding preparation for docking, or blind docking without a defined box. Not for binding affinity prediction — the confidence score ranks pose plausibility, not potency.

The English summary is being prepared. Showing the original description.

Protein language models through the EvolutionaryScale `esm` Python SDK. Generate and embed sequences with ESM3 (multimodal sequence, structure and function prompting), extract per-residue and mean-pooled embeddings with ESM C, fold sequences with ESMFold2, and run inference locally or against the Forge and Biohub hosted clients. Use this skill for protein representation learning, variant effect and mutational scanning from likelihoods, sequence generation and inpainting, structure prediction from sequence alone, and embedding features for downstream models. Also trigger on esm, ESM3, ESMC, ESM Cambrian, ESMFold2, `from esm.models`, ESMProtein, GenerationConfig, forge.evolutionaryscale.ai, biohub.ai, or ESM_API_KEY.

The English summary is being prepared. Showing the original description.

Drive X/Twitter from the CLI (read, post, DM, lists, communities, Grok, scheduling, analytics). Use when an agent needs to read or act on X/Twitter. Every command prints one JSON envelope.

The English summary is being prepared. Showing the original description.

Step 4 of the PaperOrchestra pipeline (arXiv:2604.05018). ONE single multimodal LLM call that drafts the remaining paper sections (Abstract, Methodology, Experiments, Conclusion), extracts numeric values from experimental_log.md into LaTeX booktabs tables, splices the generated figures from Step 2, and merges everything into the template that already contains Intro + Related Work from Step 3. TRIGGER when the orchestrator delegates Step 4 or when the user asks to "write the methodology and experiments sections" or "fill in the rest of the paper".

The English summary is being prepared. Showing the original description.