APIs have become the de facto standard of interaction across devices and platforms. With agentic AI, API proliferation is exorbitant. APIs are being implemented across IT teams via various tech stacks (Java, Python, MuleSoft, etc.), resulting in API sprawl.
API sprawl occurs when an organization adopts APIs across different technologies, standards, and deployments as their needs change over time. Eventually, this results in the uncoordinated proliferation of APIs. This sprawling effect creates challenges like:
- Poor visibility of available APIs
- Inability to ensure governance
- Increased management complexity
- Heightened security risks
API discovery and recommendations are crucial for maximizing the value of available APIs while addressing the challenges of API sprawl. We’ll discuss how Salesforce empowers API discovery and recommendation with Data Cloud and MuleSoft’s Anypoint Platform.
How Salesforce uses Salesforce technologies for API discovery
API discovery involves making APIs easily findable within an organization’s ecosystem, often through a catalog. Reusability refers to the practice of designing APIs that can be used across multiple applications or services, reducing development time and increasing efficiency.
Salesforce’s Digital Enterprise Technology organization developed a smart “Salesforce on Salesforce” approach to streamline the API discovery process by leveraging tools like MuleSoft Universal API Management solution and Data Cloud. This solution establishes a seamless approach to Universal API Management, serving as a source of inspiration for effective practices.
Anypoint Flex Gateway
Anypoint Flex Gateway is designed to manage and secure APIs running anywhere. Built to integrate seamlessly with DevOps and CI/CD workflows, Anypoint Flex Gateway delivers the performance required for the most demanding applications and AI agents while providing enterprise security and manageability across any environment. With introduction of Anypoint Flex Gateway, we could manage all non MuleSoft APIs in one place and mandated teams to document the non Mule APIs using Anypoint Exchange
Anypoint Exchange
Anypoint Exchange is our catalog or market place of all reusable, pre-built assets like API specifications, fragments, connectors, and custom components. Anypoint Exchange provides its functionalities as Rest APIs to list assets, fetch API specification as OpenAPI Specification (OAS), and publish documentation, to name a few
Anypoint CloudHub
Anypoint CloudHub 2.0 is a fully managed, containerized integration platform as a service (iPaaS). CloudHub supports building and deploying APIs using an API-led connectivity approach.
Anypoint API Experience Hub (AEH)
The Anypoint Experience Hub (AEH) is a platform for creating and managing personalized API portals for thriving API ecosystems and for growing engagement for API products. AEH plays a key role by simplifying API discovery and documentation, ensuring that Agentforce can easily access and share up-to-date API details, enhancing efficiency and collaboration across teams. Anypoint API Experience Hub is powered by Salesforce Experience Cloud which eases hosting of Agentforce Service Agent via Messaging for In-App and Web (MIAW) of the Salesforce Platform.
Agentforce Agent
Agentforce Agent is a proactive, autonomous application that provides specialized, always-on support to employees or customers or developers. It’s equipped with the necessary business knowledge to execute tasks according to their specific role. Use of Agentforce Agent is optional; it provides in conversational ability for searching /recommendation of APIs.
Data Cloud
AI solutions, including Agentforce, are powered by Data Cloud because Data Cloud offers robust, secure data foundation. DataCloud supports fast, trusted AI grounded in all data sources in important ways, including:
- All data types: Bring in structured or unstructured data into Data Cloud, including audio, support chats, PDFS, and more
- Real-time data: Process and react to data in seconds using Data Cloud
- Zero copy: Expand the reach of data by connecting data outside Salesforce using Data Federation
- Trusted data: Uses Salesforce secure AI architecture
Retrieval Augmented Generation (RAG) in Data Cloud is a framework for grounding large language model (LLM) prompts. By augmenting the prompt with accurate, current, and pertinent information, RAG improves the relevance and value of LLM responses for users.
Solution approach
Data Cloud Vector Database is a specialized knowledge repository. It uses vector embeddings, a form of data representation that allows for highly accurate similarity searches. We can bring the data directly into the Data cloud using the in built connector and ingestion APIs.
Let’s create a scheduler job using Anypoint CloudHub. The goal is to bring all API documentation from Anypoint Exchange into DataCloud using Anypoint Exchange APIs. Once the data is available in DataCloud, RAG kicks into action.
API recommender: Build an agent for consumers to search for available APIs in natural language, pass received user input as prompt to DataCloud using MuleSoft actions.
Data Cloud then uses advanced algorithms to perform a similarity search. This means it compares the context of the user input against the data stored in the Data Cloud Vector Database to find the most relevant results.
For example, if the search is related to an API for sending email, Agentforce sends the user input as a Query to Data Cloud searches Anypoint Exchange data ingested through scheduler job. Data Cloud retrieves relevant context. Once the search is complete, Data Cloud retrieves the specific pieces of information requested instead of being overwhelmed with raw data “like” normal db query, it brings only the most relevant, contextual information of APIs available based on the Vector scoring – which can be ordered and presented back to the user.
Data Cloud facilitates API discovery based on user context and needs, leveraging its built-in RAG capability. Combining this agentic search with Anypoint Experience Hub would establish a unified platform for locating both Mule and non-Mule APIs.
API discovery and beyond
Using Anypoint Exchange to document both Mule APIs and non-Mule APIs (via Flex Gateway), DataCloud for RAG and AEH for API Portal and Agentforce Agent streamlines API discovery process, enhances reusability, consistency, and improves the user experience.
With the release of FlexGateway for MCP and A2A, now we can manage all non Mule APIs, Model Context Protocol (MCP) servers, and Agent-to-Agent Protocol (A2A) in one place.
Want to see this solution in action or to learn more about Agentforce and MuleSoft? Attend an Agentforce World Tour near you, and don’t forget to head over to MuleSoft.com to learn about everything MuleSoft for Agentforce.
Editor’s note: This article was collaboratively written by Gopalakrishnan TV, Bharath Kumar NJ, and Nirbhay Sharma.