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.
-
tomevault-io Bundle Prompt Engineering SkillCraft effective prompts that get the best results from language models. Use when this capability is needed.
- tomevault-io Bundle Cross Agent Export
Use when exporting CoderSteroids instructions to other agent harnesses, creating AGENTS.md or CLAUDE.md files, checking cross-agent portability, or preparing a repository for Codex, Claude, Cursor, Gemini, OpenCode, or similar coding agents
- tomevault-io Bundle Dart Best Practices
General purity standards for Dart development. Use when this capability is needed.
- tomevault-io Bundle Langchain Data Handling
Load and chunk documents for LangChain 1.0 RAG pipelines correctly \u2014\ Use when this capability is needed.
- tomevault-io Bundle Cursor Prompting
Prompt shaping rules for delegated cursor-agent tasks Use when this capability is needed.
- tomevault-io Bundle QA Visual Testing
E2E visual testing using Playwright screenshot API with Vision MCP helpers for qualitative GDD compliance analysis. Use when validating shaders, materials, UI elements, and visual appearance against design specifications. Use when this capability is needed.
- tomevault-io Bundle Tuning Hyperparameters
Optimize machine learning model hyperparameters using grid search, random Use when this capability is needed.
- tomevault-io Bundle Francisco Perez Sorrosal Praxion Agent Evals
Agent Evals
- tomevault-io Bundle Aic Delegation Craft
Use when deciding which model-seat drives a piece of work in Cursor, setting up the operator's thinking-seat / execution-seat workflow, dispatching Composer subagents from an Opus session, configuring Custom Modes or the statusline for seat awareness, or recovering when an execution model misreads intent. Codifies the HOW of the AIC delegation protocol. Triggers on phrases like model routing, thinking seat, execution seat, Opus vs Composer, subagent dispatch, Custom Mode, statusline, switch back, misread recovery, delegate to AIC. Pairs with .cursor/rules/akos-aic-delegation.mdc (the WHEN); this skill is the HOW.
- tomevault-io Bundle Test Run Analysis
Analyze Freeplay test run results to surface insights, evaluation metrics, and test case details. Use when the user wants to review test run performance, compare two test runs, understand evaluation scores, or identify failing test cases. Do NOT use for executing new test runs (use run-test) or managing datasets (use dataset-management). Use when this capability is needed.
- tomevault-io Bundle Enterprise Source Management
Manages connected MCP sources for enterprise search. Detects available sources and handles source priority ordering.
- tomevault-io Bundle AI Readme
AI README 生成器 - 扫描代码并为项目生成 AI/人类共用的规则文档（AGENTS.md + .cursor/rules/ai-readme/）。当用户说 /ai-readme、@ai-readme、生成 AI 文档、生成项目规则文档、给项目加 AGENTS.md、初始化项目规则、为项目写 AI 文档时触发。 Use when this capability is needed.
- tomevault-io Bundle Opprime World Key
让 AI Agent 成为 Opprime World 的原住民——拥有 DID、土地、房屋、邮箱和 Portal。The key that opens the first metaverse where AI agents are natives. Use when this capability is needed.
- tomevault-io Bundle Load Inline Native Code
Helps write CUDA and HIP kernels using torch.utils.cpp_extension.load_inline(). Use when users want to write native GPU code (CUDA/HIP) inside a Python submission file. Use when this capability is needed.
- tomevault-io Bundle Graysurf Agent Kit Open Changed Files Review
Open Changed Files Review
- tomevault-io Bundle Graysurf Agent Kit Semantic Commit Autostage
Semantic Commit (Autostage)
- tomevault-io Bundle Oh My Openclaw
Agent orchestration framework for OpenClaw - ports oh-my-opencode patterns (Prometheus planner, Atlas orchestrator, Sisyphus executor) into OpenClaw-native constructs with category-based model routing, wisdom accumulation, and automated task completion loops. Use when this capability is needed.
- tomevault-io Bundle Build Agent System Zh Cn
AI Agent 系统设计与构建指南。广泛触发：只要用户在进行任何 Agent 相关的开发工作——构建、调试、重构或讨论 Agent、Agentic 架构、工具调用模式、多 Agent 编排、Agent 工作流、自主流水线、LLM 驱动的自动化、编码 Agent 系统，均应激活。覆盖 Codex SDK、Claude Agent SDK 及通用 Agent 设计模式。即使用户未明确说'Agent'，也应从上下文推断并激活（如工具调用循环、规划/执行模式、自主任务分解）。 Use when this capability is needed.
- tomevault-io Bundle Higashi Kota LLM Internal RAG Chat Storybook
Storybook Skill
- tomevault-io Bundle Harness Coder
Main coding agent for autonomous harness system. Continues work from previous sessions, implements one feature per session, coordinates with testing and review skills, and maintains clean handoffs via Archon. This is the workhorse of the harness system for multi-session development. Use when this capability is needed.
- tomevault-io Bundle Prp Framework
Product Requirement Prompt framework for generating comprehensive PRDs, implementation plans, and feature specifications. Use when creating technical specs, planning features, investigating issues, or documenting requirements for development. Use when this capability is needed.
- tomevault-io Bundle Civ Finish Quotes
Add a Civilization-style ceremonial quote when a substantial task is truly complete. Use this whenever the user or agent is wrapping up a real deliverable such as a feature, refactor, analysis, design doc, process change, report, or writing task, even if they do not explicitly ask for a quote; skip short replies, tiny fixes, and unfinished work. Use when this capability is needed.
- tomevault-io Bundle Iamcaominhtien Kanban Board MCP Unit Testing
Unit Testing Skill
- tomevault-io Bundle Itsdevcoffee Devcoffee Agent Skills Feedback
feedback
- tomevault-io Bundle J4flmao Agent Skills Blockchain Cryptography
Blockchain Cryptography
- tomevault-io Bundle Jackspace Claudeskillz Cloudflare MCP Server
Cloudflare MCP Server Skill
- tomevault-io Bundle Review AI Compat
AI-agent compatibility audit. Use when evaluating whether code changes are safely readable, analyzable, and modifiable by AI agents. Checks determinism, explicitness, complexity, and boundary clarity. Use when this capability is needed.
- tomevault-io Bundle Claude Code Docs
Local Claude Code documentation reference. Use when asked about Claude Code features, configuration, skills, hooks, MCP, sub-agents, plugins, or troubleshooting Claude Code issues. Use when this capability is needed.
- tomevault-io Bundle Ml RAG Foundations
Comprehensive guide to building Retrieval-Augmented Generation (RAG) applications by combining specific library skills. Use when this capability is needed.
- tomevault-io Bundle Algorithm Design
Design algorithms by modeling constraints, enumerating candidate strategies, proving correctness, and selecting data structures with explicit tradeoffs. Use when implementation success depends on algorithm choice or decomposition under unclear constraints, from vague outcome requests to concrete directives; do not use for persistence schema or deployment topology decisions. Use when this capability is needed.
- tomevault-io Bundle Java Style Guide
Style, review, and refactoring standards for Java codebases with strong typing, explicit failure handling, and maintainable architecture boundaries. Use when `.java`, `pom.xml`, `build.gradle`, or `build.gradle.kts` artifacts are created, changed, or reviewed and Java-specific quality rules must be enforced. Do not use for Kotlin-only, Android-Gradle-only, or non-JVM language style concerns unless Java artifacts are also changed. Use when this capability is needed.
- tomevault-io Bundle Rust Style Guide
Style, review, and refactoring standards for Rust codebases. Trigger when `.rs`, `Cargo.toml`, or `Cargo.lock` files are created, changed, or reviewed and Rust-specific quality rules (ownership/borrowing clarity, error handling, trait/API design, crate boundaries) must be enforced. Do not use for C/C++ toolchain conventions, shell-only changes, or non-Rust runtime style concerns unless Rust artifacts are also changed. In multi-language pull requests, run together with other applicable `*-style-guide` skills. Use when this capability is needed.
- tomevault-io Bundle Testing Mutation
Mutation-testing workflow for exposing weak assertions and missing behavioral checks. Use when tests pass but confidence is low and objective robustness evidence is needed; do not use before baseline tests exist. Use when this capability is needed.
- tomevault-io Bundle Vibe Agents
Generate AGENTS.md and AI configuration files for your project. Use when the user wants to create agent instructions, set up AI configs, or says "create AGENTS.md", "configure my AI assistant", or "generate agent files". Use when this capability is needed.
- tomevault-io Bundle Sync Agent Workbench
Synchronize, audit, or repair vendor-neutral agent-workbench files in a project. Use when a user wants every project to carry canonical AI_AGENT_GUIDE.md, thin Claude/Codex/Gemini/OpenCode entrypoints, portable prompts, and portable Agent Skills without depending on a vendor marketplace, plugin, global config, or submodule. Use when this capability is needed.
- tomevault-io Bundle Tsh Cursor Engineer
Expert in Cursor customization — designs, creates, reviews, and improves all Cursor customization artifacts (SKILL.md in agents/workflows/commands/internal, .mdc rules). Applies prompt engineering, context engineering, and AI engineering to maximize LLM effectiveness. Use when creating or improving agent skills, instructions, or command skills, auditing the Cursor customization layer, or enforcing separation of concerns across customization files. Invoke with @tsh-cursor-engineer. Use when this capability is needed.
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 prompt-engineering-skill, cross-agent-export, dart-best-practices. 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.