Course Objective & Scope
Writing custom glue code for every LLM application and internal system creates an unsustainable maintenance burden at enterprise scale. The Model Context Protocol (MCP) exposes tools, data sources and prompt templates to agents through a single open standard.
In this hands-on course participants explore the MCP architecture (hosts, clients, servers; tools, resources, prompts), build a custom MCP server in Python or TypeScript wrapping an internal API or database, implement OAuth and role-based access control, and leave with a production-ready MCP server template.
Who Should Attend?
- Backend and platform engineers exposing enterprise systems to AI applications
- AI engineers and solution architects building agent infrastructure
- API management and integration teams
- Prerequisite: API development experience with Python or TypeScript
Key Takeaways
- Explain the host, client and server roles and message flow in MCP
- Match tools, resources and prompts to the right integration scenarios
- Build a custom MCP server wrapping enterprise APIs and data sources
- Configure Streamable HTTP transport, authentication and permission boundaries
- Create a reusable internal MCP server template and security checklist
Curriculum
MCP Architecture and Core Primitives
- Host, Client and Server concepts; JSON-RPC messaging
- Tools, Resources, Prompts and Sampling capabilities
Building a Custom MCP Server
- Server setup with the Python (FastMCP) and TypeScript SDKs
- Wrapping internal REST APIs and SQL sources as MCP tools
Transport Layers and Deployment
- Local stdio versus Streamable HTTP for remote servers
- Container packaging, scaling and central MCP gateway patterns
Authentication, Authorisation and Security
- OAuth 2.1, user-scoped token forwarding and least privilege
- Defences against tool poisoning, rug-pulls and indirect prompt injection
Agent and Client Integration
- Connecting from Claude, OpenAI Agents SDK, LangGraph and IDE clients
- Logging, error handling and enterprise MCP checklist
Other Courses in This Category
-
Turning institutional knowledge, procedures and tools into modular, reusable agent skills: SKILL.md structure, progressive disclosure, templates, scripts and enterprise skill library governance.
- Live Course
- 5 modules
-
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
Get in touch
