Guide · Agent skill
How to use the AI and ML Workflow Bundle agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
The AI and ML Workflow Bundle (skill name ai-ml) is a planning and routing skill for building AI products. Once installed, it gives your agent a seven-phase checklist covering application design, LLM integration, retrieval (RAG), AI agents, ML pipelines, observability and security, and names other specialised skills to use at each step. This guide explains what it does, how to install it, how to trigger it, and what else you need.
AI and ML Workflow Bundlesickn33/agentic-awesome-skills · MIT Download the skill Quick start
- 1Download the ZIP. Download the ai-ml ZIP file from this page.
- 2Unzip into a skills folder. Unzip into ~/.claude/skills/ for all projects, or into a project's .claude/skills/ folder for one project.
- 3Or upload it in the Claude apps. Go to Settings → Capabilities → Skills and upload the ZIP.
- 4Install the referenced skills. Install skills such as ai-product, rag-engineer and langgraph separately, since the routing depends on them.
- 5Describe your AI project. Ask in plain language, for example about an LLM app or RAG system, or name a skill with Use @skill-name.
- 6Define requirements first. Set quality targets, data permissions, latency limits and failure recovery before building.
- 7Verify in your environment. Test each capability you use in your own environment before relying on it.
What an agent skill is and what this one does
An agent skill is a folder with a SKILL.md file that tells an AI agent how to handle a certain kind of task. The agent reads the file when the task matches the skill's description.
The ai-ml skill is a routing guide, not a tool. It does not contain code, scripts or connections to any service. Instead, it gives the agent a structured sequence for building AI products and, for each phase, suggests other skills to call next, such as ai-product, rag-engineer, langgraph or langfuse.
- Phase 1: AI application design
- Phase 2: LLM integration
- Phase 3: RAG implementation
- Phase 4: AI agent development
- Phase 5: ML pipeline development
- Phase 6: AI observability
- Phase 7: AI security
Install the ai-ml skill
Download the ZIP file from this page. Then choose one of two options.
For a personal setup available across projects, unzip the file into ~/.claude/skills/. For a single project, unzip it into that project's .claude/skills/ folder.
In the Claude apps, you can instead upload the ZIP under Settings → Capabilities → Skills.
Install the other skills it points to
The ai-ml skill only routes. The skills it names, such as ai-product, rag-engineer and langgraph, are not included in this file and must be installed separately. Without them, the routing has nothing to hand off to.
The source does not list where those referenced skills come from, so check the source you obtained them from.
Trigger the skill with natural requests
You do not need a special command. Describe the kind of work you are doing and the agent can match it to the skill. The skill's description covers LLM application development, RAG implementation, agent architecture, ML pipelines and AI-powered features.
The source also gives explicit prompts in the form Use @skill-name, for example Use @rag-engineer to design RAG pipeline. Those name the specialised skill directly.
- I'm building an LLM-powered support assistant. Walk me through the phases I need to cover.
- I want to add retrieval over our internal documents. Which AI skill should I use for each step?
- Set up an evaluation and observability checklist for my chatbot before we launch.
Use @rag-engineer to design RAG pipelineKnow what it needs before you start
The source says the workflow does not authorise delegation, hosted calls, deployment, background jobs or external messages. Before implementation, define task-level quality, retrieval and source checks, data permissions, latency and currency limits, and failure recovery.
You will also need to choose your own LLM, vector database or observability service. Each of these needs its own account and API keys. The file does not specify any of them.
- Referenced skills installed separately
- Accounts and API keys for any LLM, vector database or observability service you choose
Safety notes: scripts and network
According to the source, this skill runs no scripts and uses no network access. Its content is a checklist.
The file asks you to secure API keys, set data permissions and configure rate limiting, but it does not enforce any of these. Check them yourself, and verify each capability in your target environment. Unchecked checklist items are not complete just because the instruction exists.
What people use it for
- Planning a new LLM-powered application with a structured list of design steps
- Designing a RAG pipeline and choosing which retrieval and embedding skills to use
- Working out which agent framework skill fits a multi-agent or stateful workflow
- Creating an observability and evaluation checklist before launching an AI feature
- Reviewing an AI project against a security checklist covering input validation and audit logging
Frequently asked questions
Is the ai-ml skill free?
The skill's file is licensed under MIT, according to this page. The source does not mention pricing for it. Any LLM, vector database or observability service you choose will have its own terms and accounts.
What do I need to install for ai-ml to work?
The ai-ml skill references other skills, such as ai-product, rag-engineer and langgraph. These are not included in the file and must be installed separately. You will also need accounts and API keys for any services you later choose.
Where do I put the ai-ml skill folder?
Unzip the downloaded ZIP into ~/.claude/skills/ to use it everywhere, or into a project's .claude/skills/ folder to use it in one project. In the Claude apps, you can upload the ZIP under Settings → Capabilities → Skills.
Does the ai-ml skill write code or run scripts?
No. According to the source, the skill runs no scripts, uses no network and contains no code. It is a checklist that points the agent to other skills.
How is this different from just asking an AI chatbot about RAG or agents?
A general question gets a general answer. This skill gives a fixed seven-phase sequence and names the specialised skills to call for each phase, so the work follows a consistent structure. It is only useful if those referenced skills are installed.
Can I rely on this skill for production deployment decisions?
The source says outputs are not a substitute for environment-specific testing or expert review. Get expert review before making production deployment decisions, and define data permissions first, since the skill asks for them but does not enforce them.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
Choose the ai-ml skill if you are planning an LLM application or RAG system and want a structured checklist that tells your agent which specialised skill to use at each step. Skip it for simple one-off AI questions, for projects that have not installed the skills it references, or for production and sensitive-data decisions that need expert review.
Download the skill