Alfera Brain API: Workspace Memory and AI Answers

The Brain API lets your own product read from and write to your Alfera workspace brain, the same memory your worker uses inside Slack. You call it over plain HTTP with an API key, and it returns JSON or server-sent events. There are two broad things you can do with the Brain API:

You want	Call	What happens
Facts	`POST /v1/memory/search`, `POST /v1/memory/context`	A database query. No model runs on our side. Milliseconds. You prompt your own model with the result.
An answer	`POST /v1/answers`	The worker retrieves its own memory, reasons over it, and replies. Streaming, or queued and polled.

You can also teach the brain new facts with POST /v1/memory/facts. The facts go through the same capture pipeline the worker uses, so they are available for future answers and context packets. The base URL for every call is:

https://api.alfera.ai

If the agent or tool you are building speaks MCP, you can use the Brain MCP Server instead. It is the same brain, same keys, and same disclosure rules, exposed as tools over Streamable HTTP.

Quick links

Authentication

Create an API key, set scopes, and pass it as a Bearer token.

Search Memory

Query the workspace brain and return ranked facts.

Memory Context

Build a rendered memory block ready to drop into your own prompt.

Teach Facts

Submit new facts to the brain via the capture pipeline.

Answers

Run the worker on a question and stream or poll the reply.

Brain MCP Server

Mount the brain as tools in Claude Code, Cursor, or your own agent.

⌘I

​Quick links