AI Agents Need Retrieval and Skills
A practical framework for building AI agents that retrieve trusted knowledge, apply reusable skills, or combine both
Not every request needs the same kind of AI agent. Some requests need a trusted answer from a knowledge base. Some need a reusable expert skill to review, analyze, plan, or support a decision. Many real-world use cases need both.
This is the idea behind retrieval and skills. An AI agent should not treat every user request as a generic chat prompt. It should first understand what kind of work the request requires, then choose the right path.
What Is Retrieval?
Retrieval is the agent's ability to find relevant, trusted knowledge before answering. The knowledge might be hotel policies, product documentation, HR benefits, internal procedures, personal notes, cooking stories, prior decisions, or domain-specific records.
The goal is simple: do not make the AI guess what the system already knows. Retrieve the right knowledge first, then use that knowledge to answer accurately.
Example: Hotel Guest Support
A guest asks, "What time is breakfast?" This is a retrieval problem. The agent should look up the hotel's breakfast policy and return the answer. It does not need a deep reasoning process. It needs the correct fact.
Example: Personal Cooking Memory
A friend asks, "What was that beef dish Alex made at the spring dinner?" The agent should retrieve the relevant cooking story, menu note, or recipe memory. Again, the value is in trusted recall.
Retrieval is best when the answer already exists somewhere in the knowledge base and the main challenge is finding the right piece of information.
What Are Agent Skills?
Agent skills are reusable expert capabilities that help AI agents perform structured work beyond simple retrieval.
A skill may guide analysis, review, planning, reasoning, decision support, or operational workflows. Instead of asking the AI to improvise from a blank prompt, the agent applies a reusable skill designed around domain expertise and real-world patterns.
A skill can include instructions, required inputs, reasoning steps, evaluation criteria, constraints, output format, validation rules, and fallback behavior. Internally, the agent applies a stored expert skill to user-provided data, optionally combining retrieved knowledge, tools, and structured logic. But from the user's point of view, it is simply something the agent knows how to do.
Example: Review a Report for Risk
A manager asks, "Review this vendor report and identify risks." The answer is probably not sitting directly in a knowledge base. The agent needs to apply a risk review skill: summarize the report, identify operational risks, financial risks, compliance risks, missing information, and recommended next steps.
Example: Plan a Dinner Menu
A guest says, "Plan a dinner menu for six people, one vegetarian, with a lighter summer feeling." The agent may retrieve past menus and cooking preferences, but the core work is creative planning. It needs a menu design skill: season, balance, dietary constraints, pacing, prep complexity, and storytelling.
Skills are best when the user needs judgment, transformation, review, planning, or decision support. The answer is created through a process, not simply retrieved.
Why Do We Need Both?
If an agent only has retrieval, it can answer known questions but struggles with complex tasks. It can tell you what the policy says, but it may not know how to evaluate a case.
If an agent only has skills, it can generate polished reasoning but may miss the facts that make the reasoning trustworthy. It can sound smart while being disconnected from the real domain.
Real knowledge work often requires both. The agent needs trusted facts, but it also needs a way to apply those facts.
Retrieval Only
"What is the late checkout policy?" The agent retrieves the policy and answers directly.
Skills Only
"Draft a decision memo from these notes." The agent applies a memo-writing skill and produces a structured output.
Retrieval + Skills
"Can we approve late checkout for this VIP guest?" The agent retrieves the policy, checks guest context and room constraints, applies service rules, and recommends a response.
Both Retrieval and Skills Depend on Knowledge
It is important to avoid a false split. Retrieval uses knowledge directly, but skills also depend on knowledge. The difference is not knowledge versus skills. The difference is retrieving knowledge versus applying knowledge.
Retrieval finds the right facts. Skills apply expertise to those facts. In high-value domains, skills should not operate in isolation. They should operate on trusted knowledge, domain rules, examples, constraints, and real-world context.
Business Example: HR Assistant
An employee asks, "How many vacation days do I have?" That is mostly retrieval. The agent needs the employee's policy, tenure, location, and company rules.
Another employee asks, "I am moving from California to New York. What should I consider for benefits and payroll?" Now the agent needs more than one fact. It must retrieve relevant policies, identify affected areas, explain likely changes, and recommend who to contact. This is retrieval plus skills.
A manager asks, "Help me compare two candidates based on interview notes." That is primarily a skill. The system should follow a fair evaluation framework, avoid unsupported claims, separate evidence from opinion, and produce a consistent comparison.
Personal Example: Cooking Agent
A personal cooking agent can use retrieval to remember past dishes, guest preferences, stories, ingredients, and lessons learned. If someone asks, "What did we serve at the May dinner?" retrieval is enough.
But if the question is, "Design a new set menu for four guests, inspired by that May dinner but lighter and more seasonal," retrieval alone is not enough. The agent should retrieve the May dinner context, then apply a menu design skill: theme, progression, contrast, dietary fit, prep timing, and emotional tone.
That is where the agent starts to feel less like search and more like a collaborator.
Medical Monitor Example: Reviewing Data Listings
A medical monitor may ask an AI agent to review an adverse event listing or laboratory data listing. This is not just a question-answering problem. The agent may need to identify serious adverse events, Grade 3 or Grade 4 abnormalities, study drug discontinuations, ALT increases, dose-response patterns, skin reactions, stopping criteria, and potential safety signals.
That work requires a reusable review skill. The agent still needs trusted study context and data, but the value comes from applying a structured expert review pattern consistently.
In this kind of workflow, an agent skill is more than a prompt. It becomes a reusable way to encode how an expert reviews the data, what risks to look for, and how the results should be presented.
How We Can Make It Work
The architecture does not need to be complicated at first. The important part is to separate the types of knowledge from the types of work.
1. Structure Knowledge for Retrieval
First, collect the facts the agent should be able to trust. These can include FAQs, policies, documents, notes, stories, product details, study context, operational rules, and prior decisions.
Then shape the knowledge so it can be retrieved cleanly. A good retrieval item should usually have a clear title, question or topic, answer or content, source, tags, and update history. The goal is not to store everything randomly. The goal is to make the right knowledge easy to find.
2. Capture Reusable Agent Skills
Next, define the use cases where the agent needs to do more than answer a known question. These become reusable agent skills.
A skill might define the goal, required inputs, reasoning steps, evaluation criteria, constraints, output format, and fallback behavior. For example, a risk review skill should know what kinds of risks to look for and how to present them. A menu planning skill should know how to balance dishes, constraints, timing, and story.
3. Route the User Request
When a user asks something, the agent should decide what path to use:
- Use retrieval when the user needs a known fact or trusted answer.
- Use a skill when the user needs a structured review, analysis, plan, or decision.
- Use both when the work requires trusted context and expert processing.
Routing can start simple. The system can classify the request into a few known categories, retrieve relevant context if needed, select a skill if needed, then pass the right package to the AI model.
4. Give the Agent a Clear Job
The agent should not receive a vague instruction like "answer the user." It should receive a clear job: use this retrieved knowledge, apply this skill, respect these constraints, and return this kind of output.
This is how the system becomes more consistent. The AI model is still flexible, but it is no longer improvising the entire process.
5. Review and Improve from Real Usage
An AI agent improves as people use it. Retrieval misses show where the knowledge base is weak. Poor responses show where a skill is vague. Repeated questions reveal which answers should become structured knowledge. Repeated tasks reveal which workflows should become reusable skills.
Closing
Retrieval and skills are a simple way to think about AI agent design. Some requests need memory. Some need judgment. The best systems know the difference.
For i80agent, this distinction matters because a real domain is not just a pile of documents. It contains facts, rules, stories, decisions, workflows, and expertise. Retrieval helps the agent remember the domain. Skills help the agent act within it.