MoEngage introduced Merlin AI Custom Agents, a set of AI automation features for enterprise marketing teams that the company says provide visible reasoning, marketer-defined guardrails, and an open MCP architecture for connecting external AI tools to MoEngage data and actions.
MoEngage launched Merlin AI Custom Agents for enterprise customer engagement, introducing AI automation features that the company says are designed to give marketing teams more visibility and control over how automated decisions are made.
According to MoEngage’s announcement on PR Newswire, Merlin AI Custom Agents add visible agent reasoning, marketer-defined guardrails, and an MCP server intended to let tools such as Claude or ChatGPT access MoEngage data and actions. The company is positioning the release as part of its broader Merlin AI offering for customer engagement teams.
MoEngage describes Merlin AI Custom Agents as autonomous marketing agents for enterprise customer engagement. In its product update hub, the company lists the capability among its Spring 2026 releases and says the agents are connected through an AI Connector.
The core pitch is that marketers can create or use AI agents while retaining oversight of what those systems are doing. In the PR Newswire announcement, MoEngage highlights “full visibility” into agent reasoning, suggesting that teams can inspect how an agent reaches a recommendation or takes an action rather than treating the system as a black box.
That visibility is likely to matter for customer engagement teams because marketing automation often involves audience segmentation, personalization, campaign timing, and channel selection. The MoEngage materials do not provide independent performance benchmarks, but they frame explainability and control as central features of the release.
MoEngage also says Merlin AI Custom Agents include marketer-defined guardrails. Based on the company’s description, these guardrails are intended to let teams set boundaries around agent behavior before automation is applied to customer engagement workflows.
The announcement does not specify every type of guardrail available, but the concept is consistent with a growing enterprise focus on governed AI deployment: giving business teams a way to define what an AI system can and cannot do, while keeping responsibility for campaign strategy and customer interaction policies inside the organization.
MoEngage’s own overview page describes the update as part of “open, explainable AI automation.” That wording indicates the company is emphasizing not only automation, but also the ability for marketing users to understand and constrain that automation.
A notable part of the release is MoEngage’s reference to an MCP server. MCP, or Model Context Protocol, is an emerging approach for connecting AI tools to external systems and data sources. MoEngage says its MCP server can enable tools such as Claude or ChatGPT to access MoEngage data and actions.
In practical terms, MoEngage is presenting Merlin AI Custom Agents as something that can work beyond a single closed interface. The company’s materials suggest that external AI assistants may be able to interact with MoEngage capabilities through this architecture, though MoEngage’s public excerpts do not describe the complete permission model, supported actions, or deployment requirements.
For enterprise buyers, the open architecture claim may be important because many organizations are testing multiple AI assistants and do not want every workflow locked into one vendor’s front end. At the same time, any system that connects external AI tools to customer engagement data will require careful governance, security review, and access controls. MoEngage’s announcement emphasizes guardrails and visibility, but organizations would still need to evaluate how those features operate in their own environments.
MoEngage’s product update hub identifies Merlin AI Custom Agents as part of its Spring 2026 releases. The company’s dedicated overview page lists the product under Merlin AI Custom Agents and describes it as supporting open, explainable AI automation.
The announcement is another example of customer engagement platforms adding AI agent capabilities to existing marketing automation systems. MoEngage’s public materials focus less on replacing marketers and more on giving them AI-driven automation with visible reasoning, configurable constraints, and external tool connectivity.
Because the available information comes from MoEngage and its PR Newswire release, claims about the product’s capabilities should be read as company statements rather than independently verified results. The launch nevertheless shows how enterprise marketing software vendors are moving from AI-assisted content and recommendations toward more agentic systems that can connect data, decisions, and actions across customer engagement workflows.
The company is positioning the release as part of its broader Merlin AI offering for customer engagement teams.
AI agents for marketing workflows MoEngage describes Merlin AI Custom Agents as autonomous marketing agents for enterprise customer engagement.
In its product update hub, the company lists the capability among its Spring 2026 releases and says the agents are connected through an AI Connector.
Continue reading
