Jiwa and AI - Make the most of your Jiwa data with AI through MCP
Artificial Intelligence (AI) is here and a reality. AI is rapidly transforming the way businesses interact with their data. The question on the minds of many business managers is “how do we take advantage of AI in our business?”
It’s not a matter of whether you want to take advantage of AI or not. We believe your business must take advantage of AI. If you’ve not yet considered whether it’s beneficial to look at AI have a read of our article, consider the benefits; I’m sure you will agree that AI can enhance many aspects of your operations.
Jiwa is developing its AI integration capabilities that can work with a number of different AI models. Our preferred model is MS Copilot due to our strong partnership with Microsoft and the technical environments of our users. Read this article to learn more about the benefits of Copilot.
Can you give me an example of how AI might work for me?
Jiwa’s integration with AI tools can change the way your staff access information.
For example – you want to enable your employees to be able to perform ad-hoc queries on sales information:
- A user asks the AI agent: “Show me the top 10 customers by revenue this month.”
- The AI agent automatically discovers the appropriate GET /customers?sort=revenue&limit=10 endpoint.
- The results are retrieved, processed, and presented in natural language or as a chart.
This eliminates the friction between users and data, while keeping business systems secure and compliant.
The real impressive part of this, however, is the context. Subsequent queries can be made and the context is understood – so the user can then ask follow on questions such as:
“Now break that down by inventory classification.”
And then, perhaps you might want to not only read data, but make changes – so the user might then say:
“Increase the GP% of those 5 top selling products by another 3%”
And the REST API is used to increase the sell price of those products by the appropriate amount.
Another example might be in purchasing. AI might answer the question around how much stock you should order.
These are the benefits of AI that we are building into Jiwa as we progress with our AI implementation.
How It Works - From a Technical Standpoint
- API Registration via MCP
Businesses register their REST endpoints with an MCP-compatible layer, which describes the available resources, authentication methods, and parameters.
Jiwa is developing an MCP server for the Jiwa REST API. This will leverage the existing permissions already built into the Jiwa REST API – via a simple configuration. The MCP Server will be pointed to the Jiwa REST API, then nominate the Jiwa User to operate with via an API Key, and only the routes with permissions are exposed to the MCP server. This ensures the AI only accesses the information and permissions you want it to.
- MCP Connection added to AI tool of choice
Your AI agent (ChatGPT, Copilot or any tool supporting MCP) is configured with a custom MCP connector.
Just as OpenAPI provides machine-readable documentation for APIs, MCP communicates these capabilities in a way AI agents can understand.
- AI Agent Discovery and Execution
When a user queries the AI, the agent evaluates intent, matches it with an available MCP-exposed API, and executes the corresponding REST call.
- Natural Language Response
The raw API results are transformed into human-friendly answers, explanations, or even visualisations.
MCP is an emerging AI Standard
For many organisations, one of the biggest hurdles is simply connecting AI tools – such as ChatGPT and Copilot – with the systems and APIs that hold critical business information. This is where the Model Context Protocol (MCP) comes in.
MCP provides a standardised way to bridge the gap between REST APIs and AI-driven tools, making it easier for teams to query, analyse, and act on data through natural language. The Model Context Protocol is an emerging interoperability standard that defines how AI models and tools communicate with external systems. Instead of requiring every integration to be bespoke, MCP provides a consistent and extensible framework for exposing data sources, APIs, and operations to AI models.
For businesses, this means AI systems can discover and consume available services in a structured way – without needing custom glue code for each new integration.
Started by Anthropic (known for Claude AI), and adopted by OpenAI (ChatGPT), Microsoft (Copilot), Google (Gemini) and others, MCP has quickly emerged as the one standard to be embraced by all the players of the AI space.
A great analogy of what MCP is, comes from Anthropic’s own MCP project site:
“Think of MCP like a USB-C port for AI applications. Just as USB-C provides a standardized way to connect your devices to various peripherals and accessories, MCP provides a standardized way to connect AI models to different data sources and tools”
Why REST APIs Are the Perfect Fit
Most business already rely heavily on REST APIs to power internal applications, dashboards, and services. These APIs expose the business logic and data endpoints that users and applications depend on every day.
Jiwa users already have a REST API available, built into the product, exposing many possible operations on the Jiwa data.
By wrapping REST APIs with MCP, businesses can:
- Expose capabilities in a standardised way: AI agents can understand available endpoints, methods, and parameters without manual prompt engineering.
- Enable natural language querying: Users can ask questions like “What were last quarter’s sales by region?” and the AI agent can translate that into an API call.
- Support operational actions: Beyond querying, MCP can enable write operations – such as updating a debtor record, approving a purchase order, or triggering a workflow.
- Improve security and governance: MCP allows fine-grained control over what APIs and operations are exposed to the AI layer, ensuring compliance with business policies.
Benefits for Businesses
The model we have chosen to integrate AI with our systems has many benefits for businesses including:
- Faster AI adoption – Reduce time-to-value by leveraging existing REST APIs instead of building new connectors.
- Future-proof architecture – MCP creates a common standard that can evolve alongside AI tools and frameworks.
- Empowered workforce – Enable users to self-serve insights and actions through natural language.
- Operational efficiency – Automate repetitive tasks and streamline workflows by combining AI reasoning with existing APIs..
The Future of AI and Enterprise APIs
The Model Context Protocol is more than just an integration standard – it’s a strategic enabler for businesses to unlock the full potential of their ERP and IT investments. By exposing REST APIs through MCP, businesses can seamlessly connect their existing digital infrastructure with the intelligence of modern AI agents.
The result? A future where querying enterprise data, triggering workflows, and gaining insights is as simple as having a conversation.
In Summary
MCP provides the connective tissue between AI models and enterprise REST APIs, enabling secure, accessible, and powerful AI-driven operations across the business.
Whether you’re refining operations or reimagining customer engagement, Copilot is a tool that can help you work smarter, faster, and more strategically.