Six two-hour sessions

Get more from AI
in your own work

Work with Claude, Codex or open-source alternatives. Create reusable skills and small tools, then build a personal Knowledge Base around your experience.

From €8,000 · excl. VAT and applicable taxes

Specialists and managers who want to use AI in their own work

Use AI tools with your own tasks and materials. Learn how to take on more complex work, reduce manual steps and produce results you can check and use.

Give agents context, connect approved systems, create reusable skills and build a small application. Try different approaches and learn which one suits the task.

Part of the course focuses on your personal Knowledge Base. Describe the projects, people, topics and decisions in your experience and how they connect. Keep sources and check answers so the agent can draw on that experience when helping with new work.

What you will build

An agent set up for your work

A working folder, instructions and connections to systems your company permits you to use.

Skills for recurring tasks

Tested instructions for writing in your style and selecting cases, with a way to review and improve them.

A project map and search tool

Projects organised by industry, topic and year, with an application that selects relevant cases and shows the original materials.

A personal Knowledge Base

Case records, relationships between them and a procedure for adding new material.

Six sessions on your own materials

Before the course, install the tools and check access. We use Claude, Codex or open-source alternatives available to your company. Bring materials you are permitted to work with. Each two-hour session builds on the previous one.

1. How an agent worksTwo hours

Understand the difference between a model, an application and an environment, using Claude Code, Cowork and Codex. Explore what an agent can see and remember, what is lost when conversation history is compressed, and why important context belongs in files. Check where commands run and what the agent can access.

Practice: Give the agent a folder of your materials and ask it to describe your experience. Check what it noticed, missed or invented. Record working rules in AGENTS.md or CLAUDE.md.

Output: A working folder, initial instructions and a description of your experience.

2. Documents and toolsTwo hours

Read and create Word, Excel and PowerPoint files from examples. Check both the accuracy of the content and the quality of the layout. Explore agent tools, permissions, and connections to approved systems through the Model Context Protocol (MCP) and scripts.

Practice: Extract cases from your documents: the task, what you did, the outcome, industry and year. Build a table and check it against familiar examples. Connect another approved source, such as a shared drive, recorded presentations or correspondence.

Output: A table of project examples, working connections and a process for checking results.

3. Reusable skillsTwo hours

Write instructions for recurring tasks and package them as skills. Keep instructions, examples, templates and any scripts together. Learn to share skills with colleagues and improve them after checking the output.

Practice: Use your writing to create a skill that follows your style, then test it on new material. Build another skill that selects relevant cases from the table created in the previous session.

Output: Skills for writing in your style and selecting cases, plus personal rules for working with an agent.

4. Visuals and small applicationsTwo hours

Create images, editable slides, HTML pages and small applications. Provide a reference and brand guidelines. Define what an application should do, where it should store data and how to verify its behaviour.

Practice: Map your projects by industry, topic and year. Build an application that takes a request from a client or manager and finds relevant cases with source links. Check the selection for completeness and accuracy.

Output: A project map, a case-search application and a skill for keeping the cases up to date.

5. Your own Knowledge BaseTwo hours

Decide what work the Knowledge Base should help with. Describe its entities and relationships: cases, clients, industries, topics, people and decisions. Organise folders and Markdown files, add metadata, use consistent names and keep source materials separate from extracted facts.

Practice: Define three to five questions about your experience: where you solved a similar problem, what you know about an industry, or which ideas remain unfinished. Create case records with sources. Ask questions and trace the answers back to the evidence.

Output: A personal Knowledge Base structure, documented relationships and case records.

6. Keeping the Knowledge Base usefulTwo hours

Add new knowledge with sources, identify contradictions and correct errors. Configure skills for updating the Knowledge Base and explore scheduled agent work.

Practice: Add materials from the past month. Use a client or manager’s request to find relevant cases and prepare a written response and an illustration. Make a plan for working with the Knowledge Base over the next month.

Output: An update procedure, a response to a real work request and a personal action plan.

Training for your team

Six two-hour sessions

In-house, using your own materials

From €8,000 · excl. VAT and applicable taxes

We agree the group size, delivery format and dates with you.

The programme includes six two-hour sessions, setup and access checks before the start, learning materials, templates and starter skills.

Request

Discuss AI foundations

Tell us what your team needs. We’ll reply to arrange a 30-minute call and talk through the options.

We use your details to respond to your enquiry. Please avoid sending confidential information in the message.

Compare all training and implementation programmes →

Let’s
work together

Tell us what your team needs to learn, build or change. We’ll help you choose a training programme or plan implementation.

See the work

Read how a US bid firm chooses which bids to chase from its own track record, and how an executive search firm works with AI agents.

Read the cases
Kiberry

© 2026 Kiberry — the Cyprus-based practice of ScrumTrek

Operated by Skill Tech Consulting Limited, Hong Kong SAR · Registration no. 76164705