Question 1

Is Atender MCP-native?

Accepted Answer

Yes. Atender ships a first-class MCP server at v3.atender.com/api/v1/mcp — a streamable-HTTP server that drives the same API the product runs on. Connect any MCP client and the available tools load automatically, with the scopes each tool needs discoverable from the server itself.

Question 2

Which MCP clients work with Atender?

Accepted Answer

Any client that speaks the MCP spec: Claude, Claude Code, Codex, Cursor, and custom agents built on the various agent SDKs. There is no Atender-specific plugin to install — point the client at the server URL, authenticate, and go.

Question 3

Can I use the entire platform through the API?

Accepted Answer

Yes — the REST API is the same surface the product uses, not a limited export layer. Conversations, contacts, AI agents, knowledge base, cases and incidents, analytics, webhooks, and admin are all callable, with scoped keys and sandbox workspaces for safe development.

Question 4

Can I configure AI agent stacks via the API or MCP?

Accepted Answer

Yes. Agent stacks — the agents, their models, knowledge scopes, tools, and guardrails — are regular API objects. Create and update them from code, from CI, or let a coding agent do it over MCP; changes take effect immediately, no UI round-trip.

Question 5

How do coding agents like Claude Code or Codex use Atender?

Accepted Answer

They connect to the MCP server, discover the tool catalogue, and work with your workspace like a teammate: search the knowledge base, draft and publish articles, inspect conversations, configure agent stacks, and wire webhooks. Scoped keys and OAuth consent keep them inside exactly the permissions you grant.

Question 6

How does authentication work for the MCP server?

Accepted Answer

Two ways. Headless setups use a tenant API key with the scopes you choose. People connecting their own client use OAuth: the browser opens, an owner signs in and approves the requested scopes, and the client receives revocable tokens with PKCE. Either way, every tool call is bounded by the granted scopes.

Question 7

How do webhooks work, and how are they secured?

Accepted Answer

Subscriptions are created via the API and push events — conversations, cases, knowledge articles, SLAs — the moment they happen. Every delivery is signed with an HMAC secret you verify on receipt; failed deliveries retry with backoff and surface in the delivery log. No polling required.

Question 8

Are there official SDKs, or just the REST API?

Accepted Answer

Official SDKs for TypeScript/JavaScript and Python cover the full API surface, with OpenAPI specs available for generating clients in other languages. And since the MCP server exposes the platform as tools, agents often need no SDK at all.

Question 9

Is there a sandbox environment for testing?

Accepted Answer

Yes. Sandbox workspaces let you build and test against realistic data without touching production. API keys are scoped to sandbox or production explicitly, so pointing a dev script at production isn't a one-character mistake away.

Question 10

What are the API rate limits?

Accepted Answer

Standard limits cover normal usage with headroom — typically thousands of requests per minute per workspace, visible in X-RateLimit-* response headers. MCP connections share a per-workspace budget, and limits can be raised on enterprise plans for high-throughput scenarios.

Built for the agent era — MCP & API native

Every feature, callable

Point any agent at Atender

Agent stacks, configured from code

Events pushed, never polled

AI Agents

Knowledge Base

Automation

Integrations

The platform your agents can drive.

See Atender in action