Salesforce has announced its new vibe coding product, Agentforce Vibes. This programming tool allows developers to work independently on Salesforce apps and agents, automatically performing a large part of the technical tasks. Agentforce Vibes guides developers from the concept stage to the creation and monitoring of the application, while having built-in enterprise security and management mechanisms.
Salesforce has announced its new vibe coding product, Agentforce Vibes. This programming tool allows developers to work independently on Salesforce apps and agents, automatically performing a large part of the technical tasks. Agentforce Vibes guides developers from the concept stage to the creation and monitoring of the application, while having built-in enterprise security and management mechanisms.
The new tool includes a standalone AI coding agent called Vibe Codey. The agent is already connected to a company’s existing Salesforce account, allowing it to reuse the organization’s already written code and follow its coding guidelines to build apps that align with existing products, TechCrunch reports .
As Dan Fernandez, VP of Product for Salesforce Developer Services, told TechCrunch, connecting Agentforce Vibes to an existing company account provides businesses with a dual benefit: They get to dive into Vibe coding while avoiding potential security risks and having to start from scratch for each project.
“We try to provide you with everything you need,” Fernandez said. “So instead of spending a lot of time configuring [model context protocols—systems that allow AI models to securely communicate with external tools], setting up the dev environment, setting up the tools, everything is already built in and waiting for you, including the queries to the AI, to get started. And that’s really what lowers the barrier to entry.”
The new functionality is developed based on the Visual Studio Code Extension fork of the open AI agent Cline.
Fernandez noted that the company tested many different open source coding tools before choosing Cline, in particular because of its robust support for MCP, which allows AI models to securely interact with external tools and data.
