Course Objective & Scope
General-purpose models do not know your organisation's specific business rules, report formats or approval procedures, and stuffing everything into one giant system prompt is both costly and brittle. This programme teaches the discipline of packaging institutional expertise into modular, portable agent skills.
Following progressive disclosure principles, participants design a skill package combining metadata, step-by-step instructions, reference documents and helper scripts, test it on a case from their own department, and produce a reusable internal skill template.
Who Should Attend?
- Domain experts and business analysts encoding departmental workflows for AI
- AI engineers and prompt designers building agents and assistants
- Internal AI Centre of Excellence (CoE) teams
- Prerequisite: basic prompt writing experience; coding is not required
Key Takeaways
- Turn a business procedure into a modular, well-bounded agent skill
- Write skill descriptions that let agents trigger the right skill at the right time
- Combine instructions, examples, checklists and validation scripts in one package
- Structure progressive disclosure to keep context consumption low
- Establish versioning and quality standards for an enterprise skill library
Curriculum
From Monolithic Prompts to Modular Skills
- The skill concept: combining instructions, references, templates and tools
- Preserving the context budget through progressive disclosure
Skill Anatomy and Trigger Design
- Metadata (name, description) and accurate triggering criteria
- Step-by-step workflows, decision trees and output templates
Deterministic Checks and Helper Tools
- Backing LLM instructions with validation scripts and schemas
- Negative examples and boundary rules for common failure modes
Testing and Refining Skills
- Trigger accuracy and instruction-following test scenarios
- Iterative refinement against real case inputs
Enterprise Skill Library Governance
- Cataloguing, versioning, ownership and security review
- A standard skill template for cross-team reuse
Other Courses in This Category
-
End-to-end and step-level evaluation of multi-step agent systems: trajectory evals, tool calling accuracy, task completion metrics, simulated user/environment testing and CI/CD eval pipelines.
- Live Course
- 5 modules
-
Securing tool-using agents that act on external systems: indirect prompt injection defence, delegated user authorisation with OAuth/OBO, sandbox isolation, guardrails and OWASP Agentic security controls.
- Live Course
- 5 modules
-
Foundational programme on how LLM-based agents work: model, tools, memory and the control loop; how agents differ from chatbots and workflow automation, and how to pick the right first use case for your organisation.
- Live Course
- 6 modules
-
Agentic AI for executives: impact on business models, processes and organisation, function-by-function use cases, investment and ROI assessment, operational risk, governance and redesigning teams that work alongside agents.
- Live Course
- 5 modules
-
Hands-on programme on how LLMs use APIs, databases and internal systems via function calling / tool calling: tool schema design, argument validation, error handling and a working agent built on OpenAI, Claude and Gemini APIs.
- Live Course
- 6 modules
-
Method for turning business processes into agentic workflows: separating deterministic steps from LLM decisions, prompt chaining, routing, parallelisation and evaluator–optimizer patterns, implemented in n8n or Power Automate.
- Live Course
- 6 modules
Get in touch
