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.
-
j1nn0 Bundle Agent OrchestrationDelegates non-trivial engineering work to an `explorer` (investigation, research) and a `fixer` (implementation) in a Herdr session, while the current top-level agent keeps strategy and review. Use when a root cause is unclear, code or architecture is unfamiliar, an external API, SDK, or specification needs research, or an implementation spans several files — even when the user never mentions agents, delegation, or orchestration. Do not use when the current agent was itself delegated work as an `explorer` or `fixer`. Requires HERDR_ENV=1. Handle trivial, local, low-risk work directly.
- j1nn0 Skill Continuous Improvement Loop
Runs autonomous software-engineering improvement as repeated, reviewable rounds until the repository converges: each round reassesses the current state, selects one worthwhile improvement, uses agent-orchestration for investigation and implementation, verifies and commits the result, then decides from the new repository state whether another round is justified. Use when the user wants a codebase to keep improving autonomously rather than complete a fixed task list. Requires the agent-orchestration skill when a round meets the delegation criteria defined below.
-
him0 Skill CodexDelegate tasks to OpenAI Codex CLI in read-only sandbox mode. Use when the user explicitly asks to use codex, or when you need a second opinion from another AI agent for coding, search, or general tasks.
-
contrario Skill Agent OsThe operating system layer for AI agents. Routes goals to the right skills. Executes with checkpoints.
-
bang9 Bundle Whip StartSpawn whip agent sessions to handle tasks. Dispatch a single agent or assemble a small team with explicit backend, scope, and ownership.
- bang9 Bundle Rewind Analyze
Analyze a session transcript with AI and generate structured insights (prompt quality, strategy critique, key decisions, takeaways) for the rewind viewer.
-
suraj1235 Bundle Odw 7Dynamic multi-agent workflows for Zed Agent and zcode. Use when the user says "ultracode", "workflow:", "/deep-research", or wants broad multi-file work with planning, parallel execution, verification, or crash-resume.
- suraj1235 Bundle Ultracode 7
Ultracode-style dynamic workflows for Zed Agent and zcode. Use when the user says "ultracode", "workflow:", "/deep-research", or wants broad multi-file work with planning, parallel execution, verification, or crash-resume.
- suraj1235 Skill Odw 8
Dynamic multi-agent workflows for Zed Agent. Use when the user says "ultracode", "workflow:", "/deep-research", or wants broad multi-file work with planning, parallel execution, verification, or crash-resume.
- suraj1235 Skill Ultracode 8
Ultracode-style dynamic workflows for Zed Agent. Use when the user says "ultracode", "workflow:", "/deep-research", or wants broad multi-file work with planning, parallel execution, verification, or crash-resume.
-
co-labs-co Bundle Skill CreatorGuide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations. ALWAYS use Context7 MCP to pull relevant documentation when creating skills.
-
futuregerald Bundle HandoffCompact the current conversation into a handoff document for another agent to pick up.
-
artsmc-claude-dev-agents Bundle Mastra AgentsMastra Agent development guide - creation, tools, memory, networks, processors, guardrails, voice, and structured output patterns. Covers the Agent constructor ('provider/model' strings), dynamic instructions/models via requestContext, inputProcessors/outputProcessors (PII detection, moderation), structuredOutput schemas, agent.network(), and requireToolApproval HITL.
- artsmc-claude-dev-agents Bundle Agent Benchmark
Agent Library Benchmark
- artsmc-claude-dev-agents Bundle Remote Control Builder
Build remote control system for Claude Code using multi-agent team
-
thangchung Skill Coffeeshop Counter Service 2End-to-end counter: identify customer, classify intent, build order, confirm, submit. ToolSearch.Gateway-first. No direct MCP or CLI calls.
-
troykelly-codex-skills Skill Code ReviewerUse when explicitly asked to run the code-reviewer subagent or when another skill requires the code-reviewer agent card.
- troykelly-codex-skills Skill Security Reviewer
Use when explicitly asked to run the security-reviewer subagent or when another skill requires the security-reviewer agent card.
-
valec3 Bundle Maintaining Brand IdentityProvides the single source of truth for brand guidelines, design tokens, technology choices, and voice/tone. Use this skill whenever generating UI components, styling applications, writing copy, or creating user-facing assets to ensure brand consistency.
- valec3 Skill Backend Testing Phpunit
Unit testing, test structure, mocking.
- valec3 Skill Backend Testing Integration
Integration tests, database testing.
- valec3 Skill Backend Architecture Modules
Modular structure, namespace organization.
- valec3 Skill Backend Security Authorization
RBAC, permissions, access control.
- valec3 Skill Backend Deployment Configuration
Environment config, .env management.
-
packmindhub Skill Michel Create Packmind DatasetCreate a Packmind dataset
-
ab-aswini Skill AuthAuthentication & Authorization Skill
- ab-aswini Skill Fastapi
FastAPI Skill
- ab-aswini Skill Testing
Testing Skill
- ab-aswini Skill Security
Security Skill
-
open-mercato Bundle Data Model Design 2Design entities, relationships, and manage the migration lifecycle. Use when planning a data model, designing entities, choosing relationship patterns, adding cross-module references, or managing database migrations. Triggers on "design entity", "data model", "add entity", "database schema", "migration", "relationship", "many-to-many", "junction table", "foreign key", "jsonb", "add column".
-
mrsdaehin Bundle K6 Grafana ValidationValidates and analyzes k6 test results against Grafana by querying Prometheus metrics via mcp-grafana, checking SLO thresholds, diagnosing bottlenecks (percentile health, HTTP duration breakdown, root-cause patterns), and verifying dashboard panels.
-
anilcancakir Bundle Agent Creator 2Authors and audits Claude Code subagents: tool allowlist or denylist, model and effort, permission mode, memory, skill preload, worktree isolation, and the system-prompt body.
- anilcancakir Bundle Prompt Writer 2
Writes and audits prompts another Claude will execute: system prompts, subagent briefings, skill and command bodies, agent bodies, CLAUDE.md and rule files. The sibling creator skills call this one for the body text.
-
av Bundle Anneal 3Use when the user wants to systematically fix AI code slop — duplicated logic, over-engineering, silent error swallowing, convention drift, cargo-cult patterns, and other LLM-introduced architectural decay — over a specified duration
-
peterfei Bundle Backend Dev Agent 2后端开发 Agent — API 设计、数据库优化、服务器端逻辑、认证授权、性能优化
- peterfei Bundle QA Engineer Agent 2
QA 工程师 Agent — 测试策略、测试用例设计、自动化测试、性能测试、安全测试
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 agent-orchestration, continuous-improvement-loop, codex. 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.