Start with the places customers already use rather than enabling every available surface at once.
Customer channels
Multichannel AI agent with one controlled identity
A multichannel AI agent lets customers reach the same owned Agent across the currently supported messaging surfaces instead of rebuilding its knowledge and behavior for every inbox.
For the usual MCP path, model-token cost to the Agent creator is $0: the customer's AI client supplies the model under that customer's subscription and limits.
Early access
What is getting in the way
Each channel becomes its own small support system. Answers drift, credentials scatter, and customers get a different experience depending on where they ask.
What changes
Configure one published Agent with an explicit channel model, then connect only the channels whose credentials, policies, and customer identity rules you can support.
Customer channels
How Prostir gets you there
Choose the real channels
Start with the places customers already use rather than enabling every available surface at once.
Configure one Agent runtime
Select the Agent's channel model, confirm provider or hosted-credit readiness, and keep the Agent-level Channels switch explicit.
Connect and verify each surface
Add provider credentials and settings, enable the channel, and test inbound identity, replies, failures, and any customer authorization.
Early access
What you can verify
- Shipped channel kinds are Telegram, Slack, email, WhatsApp, Instagram, Messenger, Microsoft Teams, and the web widget.
- Inbound messages run through the selected published Agent and return on the same configured surface.
- Every channel remains gated by publication, feature and plan state, a valid model selection, enabled credentials, and provider-specific requirements.
A good fit if
- Your customers already contact you through two or more currently supported surfaces.
- You want one owner-controlled Agent and knowledge boundary behind those conversations.
Not the right fit if
- You need every provider feature, unsupported identity method, or arbitrary messaging network.
- You need channel-native checkout or subscriptions that have not been proven for that exact Store and channel.
Questions before you decide
01Does turning on Channels make every provider work automatically?
No. Each surface needs its own supported credential and settings, while the Agent also needs a valid channel model, publication state, feature access, and any required credits.
02Can customers keep the same private memory everywhere?
Only when the channel can establish an allowed customer identity and grant. Unsupported proof methods remain gated; anonymous or cross-Agent shared memory is not implied.
03Who pays for model tokens, and what do they cost the Agent creator?
For the usual MCP path, model-token cost to the Agent creator is $0. When a customer uses the Agent through a supported MCP connector in ChatGPT or Claude, the model runs in that customer's AI environment under the customer's subscription and limits. The creator pays for the Prostir plan, not those model tokens. AI Credits are needed only when the creator separately chooses a Prostir-hosted model run, such as a supported channel or widget. BYOK is optional and bills the creator's own provider account. Paid external tools remain separate.
Explore related options
Early access
Pick the channels that matter first
Tell us where customers ask, how they identify themselves, and what the Agent should handle. We will map supported channels and the setup each one requires.