AI & ML
AI & ML agent skills cover the machine-learning workflow itself: writing and evaluating prompts, building RAG pipelines, running evals, and wiring up model APIs. Each one is a SKILL.md file your agent loads on demand, so the know-how travels across Claude Code, Cursor, and 60+ agents.
-
h1d3r Skill CsharpApply project-aware C# conventions for applications, libraries, APIs, and tests.
- h1d3r Skill Go Pro
Design production Go services with clear package boundaries, errors, observability, and performance evidence.
- h1d3r Skill Python
Apply project-aware Python conventions for services, libraries, scripts, refactors, and validation.
- h1d3r Skill Backend
Design or modify backend routes, services, repositories, configuration, and error handling with clear boundaries.
- h1d3r Skill Testing
Plan focused unit, integration, contract, and end-to-end testing with reproducible evidence.
- h1d3r Skill Rust Pro
Handle advanced Rust ownership, lifetimes, traits, unsafe code, FFI, and measured performance work.
- h1d3r Skill Security
Review and implement authentication, authorization, secrets, input handling, privacy, and compliance controls.
- h1d3r Skill API Design
Design APIs and integrations with explicit consumers, resources, errors, idempotency, and lifecycle decisions.
- h1d3r Skill Csharp Pro
Use modern C# typing, nullable analysis, async, dependency injection, and performance practices.
- h1d3r Skill Python Pro
Use modern Python tooling, typing, async, packaging, and evidence-based performance practices.
- h1d3r Skill Rust Async
Design Rust async tasks, cancellation, timeouts, backpressure, and Tokio lifecycle behavior.
- h1d3r Skill API Contract
Define and evolve API schemas, compatibility, authentication, errors, pagination, and examples.
- h1d3r Skill Quality Gates
Select proportionate validation gates and report commands, results, impact, and remaining risk.
- h1d3r Skill Task Template
Structure multi-step engineering work with a single active item, focused validation, and explicit delivery notes.
- h1d3r Skill Dotnet Backend
Design ASP.NET Core and EF Core services with clear layers, validation, security, and background work.
- h1d3r Skill Go Concurrency
Design and review Go concurrency, cancellation, backpressure, goroutine ownership, and shutdown behavior.
- h1d3r Skill Python Testing
Plan and implement Python tests with focused unit, integration, and regression coverage.
- h1d3r Skill Coding Standards
Apply cross-language readability, naming, error handling, typing, and minimal-change standards.
- h1d3r Bundle Project Structure
Classify and place new source, tests, docs, scripts, migrations, generated files, and runtime artifacts without polluting a repository.
- h1d3r Bundle Agent Runtime Compat
Adapt AGENTS.md and skill behavior across Codex, Cursor, Pi, GPT-6 Astra, and unknown future agent runtimes by detecting capabilities instead of guessing from model names.
- h1d3r Skill Database Performance
Diagnose and improve database performance using query plans, indexes, measurements, and safe rollout.
- h1d3r Skill Javascript Typescript
Apply project-aware JavaScript and TypeScript conventions for applications, libraries, and services.
- h1d3r Skill Python Testing Patterns
Apply pytest fixtures, mocks, parametrization, async tests, and property-based testing safely.
-
jnzader Skill Skill RecipesSelf-installing markdown recipes for skills. Each skill can include an INSTALL.md that an agent reads and executes to set up dependencies, MCP servers, or config. Trigger: When installing a skill with INSTALL.md, or creating skills with setup requirements.
- jnzader Skill Webapp Testing
Browser-based UI verification using Playwright. Page Object Model, selector best practices, visual regression, network interception, and MCP integration. Trigger: When writing E2E tests, verifying UI changes, or setting up Playwright.
- jnzader Skill Add Prompts Endpoint
This skill covers the integration of various prompt types into the system. Trigger: When working with prompts in the application.
-
harpreetsingh Skill Hs Sw Sprint GoLaunch multi-agent sprint from execution plan — spawn director who owns the team
-
valorisa Bundle Dual LLM Contradiction DialogOrchestration d'un dialogue structuré entre deux LLMs (A et B) autour d'un même sujet, avec cadrage d'intention, génération d'angles morts, recherche systématique de contradictions et garde-fous contre le contournement silencieux.
- valorisa Bundle Collaborative Technical Peer Review
Orchestration d'une revue technique collaborative multi-étapes destinée à transformer une analyse initiale en un artefact technique robuste (documentation, article, dépôt GitHub, spécification ou rapport).
-
richlander Skill MarkoutUse when generating Markdown or other structured output (plain text, ANSI, pretty tables, TSV/JSONL) from C# objects instead of hand-built strings — CLIs, tools, reports, agent output. Markout is a source-generated .NET serializer: it looks like System.Text.Json source-gen but the rules differ (NO reflection fallback), so it needs a generated MarkoutSerializerContext and Markout-specific attributes. Start here for the required pattern; branch to the domain skills for conditional reports, multi-view/verbosity, output formats, built-in shapes, and composite cells/cards. Don't decompile the Markout assembly or web-search its API — every idiom you need is in these skills.
- richlander Skill Output Formats
Use when you need output other than default Markdown — plain text / Unicode, ANSI terminal (Spectre), pretty aligned tables, or TSV/JSONL exports — or when one model must serve several formats from a single render path (a CLI `--format` switch). Pick the format with a formatter + MarkoutWriterOptions; never hand-build TSV/JSONL. Requires the base `markout` pattern. Don't decompile the assembly or web-search the API — the TSV/JSONL/formatter idioms are all here.
- richlander Skill Built In Shapes
Use when a report needs rich visual elements — bar charts, stacked/proportional bars, alert boxes, tree hierarchies, term/definition glossaries, or code blocks — instead of hand-drawn ASCII or manual Markdown. Markout ships these as data types (Metric, Breakdown/Slice, Callout, TreeNode, Description, CodeSection) you attach as model properties. Requires the base `markout` pattern. Don't decompile the assembly or web-search the API — the shape types are here.
- richlander Bundle Nuget MCP
Experiment harness shell that attaches the real upstream NuGet MCP server (NuGet.Mcp.Server). Carries no inline grounding; package context is delivered only via the server's get_package_context tool when the agent chooses to call it.
- richlander Skill Composite Cells Cards
Use when a single value must render dense in Markdown but decompose into typed columns in TSV/JSONL — before/after changes, fractions, shares, percentages, segment breakdowns, deltas and goal polarity — or when building metric / role-matrix / verdict cards from one model (Change<V>, Fraction, Share, Percent, Segments, Delta/Goal, MetricChange<T>, MultiSourceRow, Verdict). This is Markout's most advanced tier. Requires the base `markout` pattern. Don't decompile the assembly or web-search the API — these composite types are here.
- richlander Skill Conditional Composition
Use when a Markout report must adapt to its data — show or hide whole sections, drop table columns that are empty/uniform, filter output to a chosen set of sections, or render one model into several shapes ("one model, many views"). This is Markout's highest-value idiom: declare the conditions with attributes instead of hand-writing if/else + StringBuilder. Requires the base `markout` pattern (model + context + Serialize). Don't decompile the assembly or web-search the API — the conditional idioms are here.
- richlander Bundle Markout Realmcp
Experiment harness shell: attaches the real NuGet.Mcp.Server to the Markout task. Carries no inline grounding; package context is delivered only via the real server's get_package_context tool (README or the package's shipped agent doc, depending on what ships in the package).
Frequently asked questions
What are AI & ML agent skills?
AI & ML agent skills cover the machine-learning workflow itself: writing and evaluating prompts, building RAG pipelines, running evals, and wiring up model APIs. Each one is a SKILL.md file your agent loads on demand, so the know-how travels across Claude Code, Cursor, and 60+ agents.
Which AI & ML skills are most installed?
Popular AI & ML skills on SkillMD right now include csharp, go-pro, python. Rankings shift as installs change; sort this page by "Most installs" for the live list.
Do AI & ML skills work with Claude Code and Cursor?
Yes. Every skill here ships as a SKILL.md file, an open format that works in Claude Code, Claude.ai, Cursor, Codex, Windsurf, and 60+ other agents. Install one with npx skillmds@latest add <owner>/<name>, or copy the file into your agent's skills directory.