Getting Started with the Mind Reasoner API

Step 1: Create Your First Mind

First, create a “mind.” This is the core entity that will hold the data and personality model. Send a POST request to the /minds endpoint with the name of the mind you want to create.

$
curl -X POST https://app.mindreasoner.com/api/public/v1/minds \
>
  -H "Authorization: Bearer {{API_KEY}}" \
>
  -H "Content-Type: application/json" \
>
  -d '{"name": "{{MIND_NAME}}"}'

The API will respond with the new mind’s data. Save the id from the mind object and the id from the digitalTwin object. You will need them for the next steps.

1
{
2
  "mind": {
3
    "id": "123e4567-e89b-12d3-a456-426614174000", // <-- Save this mindId
4
    "name": "Example Mind",
5
    // ...
6
  },
7
  "digitalTwin": {
8
    "id": "abcde123-e89b-12d3-a456-426614174001", // <-- Save this digitalTwinId
9
    "name": "Default",
10
    // ...
11
  },
12
  // ...
13
}

Step 2: Upload a Data Snapshot

Now, let’s give the mind some knowledge by uploading a data file (like a conversation transcript in .vtt, .docx, or .pdf format). This creates a snapshot of the mind’s state.

$
curl -X GET "https://app.mindreasoner.com/api/public/v1/minds/{{MIND_ID}}/signed-url" \
>
  -H "Authorization: Bearer {{API_KEY}}"

1
{
2
  "signedUrl": "https://storage.googleapis.com/...", // <-- The temporary upload destination
3
  "fileName": "...",
4
  "artifactId": "fedcba98-e89b-12d3-a456-426614174002" // <-- Save this artifactId
5
}

$
curl -X PUT "{{SIGNED_URL}}" \
>
  -H "Content-Type: text/vtt" \
>
  --data-binary "@/path/to/your/transcript.vtt"

$
curl -X POST "https://app.mindreasoner.com/api/public/v1/minds/{{MIND_ID}}/snapshots" \
>
  -H "Authorization: Bearer {{API_KEY}}" \
>
  -F "digitalTwinId={{DIGITAL_TWIN_ID}}" \
>
  -F "artifactId={{ARTIFACT_ID}}"

1
{
2
  "message": "Mind assessment started",
3
  "mindAssessmentId": "98765432-e89b-12d3-a456-426614174003", // <-- Save this snapshotId
4
  "artifactId": "fedcba98-e89b-12d3-a456-426614174002"
5
}

Step 3: Run Your First Simulation

Snapshot processing is asynchronous. Before you can run a simulation, you must check its status.

$
curl -X GET "https://app.mindreasoner.com/api/public/v1/minds/{{MIND_ID}}/snapshots/{{SNAPSHOT_ID}}/status" \
>
  -H "Authorization: Bearer {{API_KEY}}"

1
{
2
  "id": "98765432-e89b-12d3-a456-426614174003",
3
  "status": "completed",
4
  // ...
5
}

Once the snapshot is complete, you can run a simulation. Send a POST request to the /simulate endpoint with your **mindId** and a scenario.

$
curl -X POST https://app.mindreasoner.com/api/public/v1/simulate \
>
  -H "Authorization: Bearer {{API_KEY}}" \
>
  -H "Content-Type: application/json" \
>
  -d '{
>
        "mindId": "{{MIND_ID}}",
>
        "selectedSimulationModel": "{{SIMULATION_MODEL}}",
>
        "scenario": {
>
          "message": "What would be your response to this customer complaint?"
>
        }
>
      }'

The API will return the simulated response from the mind.

1
{
2
  "message": "Based on the provided data, my response to the customer complaint would be..."
3
}