> ## Documentation Index
> Fetch the complete documentation index at: https://dialnexa.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Get Memory

> Use the get_memory MCP tool on the DialNexa remote MCP server.

<span data-api-safety-label="read-only"><Badge color="gray" size="sm" shape="pill">Read only</Badge></span>

The organization's assistant memory, ranked for a query: the confirmed BUSINESS PROFILE (company, customers, brand voice, rules), recaps of EVERY earlier assistant conversation (agent editors and the dashboard assistant), and the DECISIONS recorded on agents (accepted/rejected proposals with reasons). Narrow with `agent` (a name like "Tanya" or an agent\_ id) and/or `query` (a keyword like "Hinglish" or "pricing"). Use for "what did we decide about X on agent Y", "which agents changed this week", "what is our policy on pricing", "does this match our brand voice". The profile describes ONE business and the result names it: never apply it to an agent that serves a different business, and if "my business" could mean either, ask which. Not call transcripts — this is the editing history and the business profile, not the calls.

## Parameters

| Name    | Type     | Required | Description                                                                                                           |
| ------- | -------- | -------- | --------------------------------------------------------------------------------------------------------------------- |
| `agent` | `string` | No       | Agent name or agent\_ id. Omit to search every conversation in the organization, including dashboard-assistant chats. |
| `query` | `string` | No       | Keyword to filter recaps and decisions. Omit for the most recent.                                                     |

## Complete input schema

```json theme={null}
{
  "type": "object",
  "properties": {
    "agent": {
      "type": "string",
      "description": "Agent name or agent_ id. Omit to search every conversation in the organization, including dashboard-assistant chats."
    },
    "query": {
      "type": "string",
      "description": "Keyword to filter recaps and decisions. Omit for the most recent."
    }
  },
  "additionalProperties": false,
  "$schema": "http://json-schema.org/draft-07/schema#"
}
```

## Registered output schema

```json theme={null}
{
  "type": "object",
  "properties": {
    "text": {
      "type": "string",
      "description": "Business facts, recaps and decisions grouped by source and agent, most relevant first, or an empty string when nothing matched"
    },
    "facts": {
      "type": "number",
      "description": "Number of confirmed business-profile rows matched"
    },
    "episodes": {
      "type": "number",
      "description": "Number of session recaps matched"
    },
    "decisions": {
      "type": "number",
      "description": "Number of ledger lines matched"
    },
    "floor": {
      "type": "boolean",
      "description": "True when nothing matched the query and the most recent entries were returned instead"
    }
  },
  "required": [
    "text",
    "facts",
    "episodes",
    "decisions",
    "floor"
  ],
  "additionalProperties": true,
  "$schema": "http://json-schema.org/draft-07/schema#"
}
```

## Before you call

* Connect and initialize the MCP client with OAuth or an API key for the intended workspace.
* This tool has no required arguments.

## Example MCP request

Send `tools/call` after your client completes MCP initialization. Replace placeholder values with IDs and inputs from your workspace. For object and array parameters, follow the nested requirements in the parameter description instead of treating an empty value as complete.

```json theme={null}
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_memory",
    "arguments": {}
  }
}
```

Successful results include `structuredContent` and a text content block containing the same JSON. Check `isError` first, then read `structuredContent` or parse `content[0].text`. See [MCP responses and errors](/docs/mcp-tools/errors) for the full envelope and failure behavior.

## Verify the result

Confirm the response is not marked `isError`, parse `content[0].text` as JSON, and verify that returned IDs and workspace data match the request.

## Retry safety

This tool does not change workspace data. Retrying after a transient transport failure is safe.

## Related guides

* [Connect to the DialNexa MCP server](/docs/mcp-tools/overview)
* [Agents MCP tools](/docs/mcp-tools/agents/overview)
* [MCP responses and errors](/docs/mcp-tools/errors)