Today we're launching SuprSend Agent.
An AI agent inside the SuprSend dashboard that builds, configures, and debugs your notification setup from simple prompts.
This is part of a larger shift.
SuprSend is now an AI-first notification infrastructure, where you operate the platform by describing what you need instead of learning every know and how.
The Agent is connected to the full SuprSend API surface, the product documentation, the agent skills, and domain knowledge of how notifications actually work. So it does not just call endpoints. It understands your use case, picks the right tools, applies the correct configuration, and tests the result on your behalf.
Why we built it
Notifications are deceptively deep. A single message reaches a user through several layers of logic, and each layer has to be configured, tested, and iterated to get the right message to the right user, on the right channel, at the right time. That depth is what makes SuprSend handle real-world use cases, and it is also what takes time to learn and operate.
SuprSend Agent changes where that effort goes.
For a new user, it removes the upfront learning curve. Instead of spending the first few days building a mental model of how the platform fits together before configuring anything, you ask the Agent. It explains concepts, answers questions against the docs, and configures your first workflows for you, so you reach a working setup as soon as possible.
For an existing user, it removes the day-to-day friction. No moving across screens and clicking through the platform to get every task done. Ask the Agent to set up a workflow, design a template, debug a notification, or pull an analytics answer, and it does the work in the conversation.
What you can do with SuprSend Agent
Ask anything
Ask the agent on how to model your data and concepts, how a feature works, what an error means, what the platform best practices are, or how to approach an integration, and it responds with context from SuprSend's documentation without you opening another tab.
"How do preferences cascade from defaults to tenants to end users?"
Manage all assets
Every SuprSend capability is the Agent's to work with. It can create, retrieve, update, and commit across tenants, objects, preferences, vendors, translations, and schemas, all from the same conversation. You stop clicking between screens to set up a vendor, check an object, or update a preference category. You describe the change, and the Agent makes it on the right resource with the correct parameters.
"Update the SMS vendor for tenant cypher and show me the current preference categories."
Design and test notifications
Building a notification normally means making every decision yourself: how the flow should be structured, when each message should go out, what should happen if a user does or does not act, which channels to use, and how the content should look and read for each one. Each choice takes product knowledge, and getting it wrong costs a rebuild.
With the Agent, a prompt like "set up a trial expiry reminder for new users" is enough.
It maps out the workflow nodes, sets the time delays, adds the branching logic, picks the right channels, designs and personalizes the templates with variables, and tests the result across every channel. You describe the outcome you want, not the build. The Agent handles the build because it knows SuprSend's primitives and how a good notification is structured.
Check analytics
A fixed dashboard can only answer the questions it was built for.
The Agent reads your notification data through a semantic layer over the data lake, so you can ask anything: how a workflow is performing, which notifications are lowest performers in last 30 days, how channels compare across segments and tenants, how many workflows are active, which is the best performing template in last 7 days, and much more.
It queries and returns an answer, helps you visualize the data as charts, and can post the result straight to Slack.
"Show me delivery and open rates for the weekly-digest workflow, split by tenant, for the last 30 days."
Debug and inspect delivery
When a notification does not deliver - the cause could sit at the channel, a preference opt-out, a tenant override, or a vendor failure, and finding it means searching through logs and screens. The Agent traces the delivery, finds the issue that stopped it, explains the cause and potential solution.
"User john@cypher.com says he is not getting emails. What is wrong?"
Safe to run on your live workspace
The Agent acts on real organizational data, so the safety model is built in at every step.
- Workspace-scoped access. Every session is scoped to a single workspace. A conversation in one workspace cannot read or change data in another.
- Service token authentication. Every session authenticates with a workspace-scoped JWT that expires and auto-refreshes within the session, so access is short-lived by default.
- Permission before changes. The Agent asks for confirmation before any edit. Update actions also read a resource's current state before changing it, so an edit never silently overwrites existing data.
- No destructive deletes. Deletes are not exposed to the Agent at all. Only low-risk, reversible operations, like removing a user from a list, are permitted, so the Agent cannot accidentally delete something it should not.
- Validated, not guessed. The Agent runs on validated SuprSend schemas and product knowledge, so it uses the correct parameters and structures instead of guessing them.
Who it is for
SuprSend Agent is for the developers, product managers, marketing, customer success, and operations teams who manage notifications but do not write the integration code. The work that used to wait on engineering now happens in a conversation on the dashboard.
Try SuprSend Agent now or read the docs to start with your first prompt.
