canonical: https://jentic.com/apis/moveo.ai/moveo # Moveo.AI The Moveo.AI API builds, trains, and runs conversational AI agents, which Moveo calls brains, and connects them to channels such as Facebook, Viber, and a web widget. Brains are configured with intents, entities, dialogs, insights, webhooks, and a model strategy, then versioned, published, and retrained. Environments, called desks, hold sessions, conversations, routing rules, departments, SLAs, business hours, macros, and channel integrations for the live support team. Knowledge bases ingest documents from datasources and can be searched, while contacts, campaigns, and reports cover customer records, outbound messaging, and analytics. ## For AI agents Send messages to Moveo.AI conversational agents, manage their intents, dialogs, and knowledge bases, and run desks with sessions, conversations, routing, and campaigns. ## Scope Does not process payments, host phone numbers, or run a ticketing help center. Use for Moveo.AI conversational agents, desks, knowledge bases, and contacts only. ## Capabilities - Open a session on a desk, send it a message with input and context, and read the session transcript - Train a brain from intents, entities, and dialogs, then create, diff, and publish versions - Search a knowledge base by natural-language query and publish or draft its documents - Connect datasources to a knowledge base and track their sync status - Route live conversations with desk rules, departments, SLAs, business hours, and capacity policies - Send outbound messages and ephemeral or agentic campaigns to customer contacts - Read contact activity, recaps, memories, and debts for a customer record ## Use cases ### Agent that talks to a Moveo virtual assistant on a customer's behalf An orchestrating agent can hand a customer question to a Moveo virtual assistant without a developer wiring the Authorization header, the right regional host, and session handling by hand. Through Jentic the agent searches for 'send a message to a Moveo session', loads the session message operation with its input schema, and sends the input and context. Your own Jentic One instance attaches the key at call time, and the reply comes back from the brain configured on that desk. Example prompt: Call POST `/v1/desks/{desk_id}/sessions` to open a session, then POST `/v1/desks/{desk_id}/sessions/{session_id}/message` with input and context, and return the assistant's reply ### Improving a brain from real conversations Conversation designers refine a brain by adding intents and entities, generating an intent from sample expressions, and reviewing and flagging the recommendations Moveo returns for the brain. Once changes are ready, a retrain is queued, the new version can be diffed against the previous one, and the version is published. Example prompt: Call POST `/v1/brains/{brain_id}/intents` to add an intent, POST `/v1/brains/{brain_id}/train` to queue a retrain, then POST `/v1/brains/{brain_id}/publish` ### Grounding answers in a knowledge base Support assistants answer from company documents held in Moveo knowledge bases. Datasources connected to a knowledge base sync documents in, drafts can be published to make them searchable, and a search call returns up to 20 matching documents for a natural-language query. Example prompt: Call POST `/v1/collections/{collection_id}/search` with query 'how do I return an item' and limit 5, and summarise the top documents ### Monitoring a live support desk Support leads need a view of workload across a desk. Conversation counts by status, by agent, and by department show where queues are building, and individual conversations can be fetched or updated. Routing rules, SLAs, and business hours can be adjusted on the same desk. Example prompt: Call GET `/v1/desks/{desk_id}/conversations/counts` and GET `/v1/desks/{desk_id}/conversations/department-counts`, then flag any department whose active count exceeds 20 ## Key endpoints | Method | Path | Description | | --- | --- | --- | | POST | `/v1/desks/{desk_id}/sessions` | Create a session | | POST | `/v1/desks/{desk_id}/sessions/{session_id}/message` | Send message to a session | | GET | `/v1/desks/{desk_id}/sessions/{session_id}/transcript` | Get session transcript | | POST | `/v1/collections/{collection_id}/search` | Search a knowledge base | | POST | `/v1/brains/{brain_id}/train` | Queue a brain retrain | | POST | `/v1/brains/{brain_id}/publish` | Publish a brain version | | GET | `/v1/desks/{desk_id}/conversations` | List conversations | | POST | `/v1/desks/{desk_id}/conversations/outbound` | Send outbound message | ## Key resources - **Brains** — Conversational AI agents with intents, entities, dialogs, insights, webhooks, versions, and model strategy - **Desks** — Environments with sessions, conversations, integrations, rules, departments, SLAs, macros, and views - **Collections** — Knowledge bases with documents, datasources, publish and draft states, and search - **Contacts** — Customer records with identities, activities, recaps, memories, and debts - **Campaigns and reports** — Outbound campaigns, including agentic campaigns, and analytics reports ## AI readiness This API is usable in Jentic One now. Its AI-readiness score against Jentic's framework shows where it stands today and where improvements would make it even easier for agents to use. - **Score:** 55 / 100 - **Maturity:** Foundational - **Dimensions:** - Foundational Compliance: 47 / 100 - Developer Experience & Jentic Compatibility: 76 / 100 - AI-Readiness & Agent Experience: 55 / 100 - Agent Usability: 42 / 100 - Security: 63 / 100 - AI Discoverability: 73 / 100 - **View full report:** https://jentic.com/apis/moveo.ai/moveo/scorecard - **How the score is calculated:** https://docs.jentic.com/reference/api-readiness-framework/overview/ - **More about the dimensions:** https://docs.jentic.com/reference/api-readiness-framework/specification/#dimensional-model-overview ### Score it yourself Every API in the directory is allowlisted, so you can re-score it with no key required. - **Score your own API:** https://jentic.com/scorecard.md - **Scoring CLI agent skill:** https://github.com/jentic/jentic-api-scorecard/blob/main/skills/jentic-api-scorecard/SKILL.md ```sh npx @jentic/api-scorecard-cli score ``` ## Why Jentic - **Setup:** Wiring Moveo.AI by hand means choosing a JWT or an API key of the right type, picking the Europe, United States, or Brazil host for your account, and threading desk, brain, and session ids through every call. Through Jentic you install once, import Moveo.AI from the API Directory, store the key once, and your agent calls it. - **Permission scoping:** Moveo puts the desk_id, brain_id, and collection_id in the URL path, as in `/v1/desks/{desk_id}/sessions`, so a rule can hold your agent to one desk or one brain and the operations you pick. You choose the operations it may call, so destructive ones like deleting a brain or a conversation are not included unless you add them. - **Credential handling:** Your Moveo.AI key is stored once, encrypted, by your own Jentic One instance and injected at execution time. It never enters the agent's prompt, logs, or context. - **Discovery method:** Agents search Jentic by intent such as 'send a message to a Moveo session' or 'search the knowledge base', and Jentic returns the matching Moveo.AI operation with its input schema out of the full operation set. ## Related APIs - **Intercom API** — Customer messaging with conversations, contacts, and help center - **Zendesk API** — Help desk ticketing with support agents and macros - **OpenAI API** — Language models for drafting replies and summarising transcripts ## FAQ ### What authentication does the Moveo.AI API use? Per its OpenAPI spec, the Moveo.AI API accepts either a JWT bearer token or an API key sent in the Authorization header. API keys come in two types: manage keys for brains, broadcasts, campaigns, collections, desks, plugins, and roles, and use keys limited to sessions, message, and classify. Through Jentic the key is stored by your own Jentic One instance and attached at call time. ### Which regional host should I call for the Moveo.AI API? The spec lists three servers: https://api.moveo.ai/api for Europe, https://api.us-central.moveo.ai/api for the United States, and https://api.sa-east.moveo.ai/api for Brazil. Use the host for the region your Moveo account lives in. ### What are the rate limits for the Moveo.AI API? The OpenAPI spec does not specify rate limits. Check the Moveo documentation at https://docs.moveo.ai for current limits. ### Is there a Moveo.AI MCP server? You don't need an MCP server to give your agent Moveo.AI. Jentic connects it directly from the API Directory: import it, store your API key once, and your agent calls session, knowledge base, and desk operations on demand without another server's tool definitions in its context. ### Can I limit what my agent is allowed to do with the Moveo.AI API? Yes. Write a rule that allows only the session create, session message, and transcript operations under one desk_id, so the agent can hold conversations on that desk but cannot delete brains, change routing rules, or launch campaigns. Every call it makes is logged. ### How do I search a Moveo knowledge base through Jentic? Search Jentic for 'search a Moveo knowledge base', load the POST `/v1/collections/{collection_id}/search` operation, and execute it with a query and an optional limit up to 20. To run it on your own infrastructure, install Jentic One from its GitHub repo.