Quickstart

First Memory in 3 Minutes

Store and recall your first AI memory. Three steps. Two curl commands. Zero dependencies.

1

Get your API Key

Sign in to your dashboard and generate a Connect Token. Copy the sk_connect_... token — you'll need it for the next steps.

2

Store a Memory

Replace sk_connect_YOUR_KEY with your token and run:

bash
curl -X POST https://forge.synapselayer.org/api/v1/capture \
  -H "Content-Type: application/json" \
  -H "x-connect-token: sk_connect_YOUR_KEY" \
  -d '{"content": "My first memory — the user prefers dark mode", "source": "quickstart"}'

Expected response:

json
{
  "status": "stored",
  "id": "clx1abc...",
  "tq": 0.4533,
  "pipeline": {
    "sanitized": false,
    "privacyApplied": true,
    "intent": "preference",
    "piiRedacted": 0
  },
  "source": "quickstart",
  "elapsed_ms": 1250
}
3

Recall the Memory

Query your memory using natural language:

bash
curl -X POST https://forge.synapselayer.org/api/v1/recall \
  -H "Content-Type: application/json" \
  -H "x-connect-token: sk_connect_YOUR_KEY" \
  -d '{"query": "user preferences", "limit": 5}'

Expected response:

json
{
  "memories": [
    {
      "id": "clx1abc...",
      "encryptedContent": "...",
      "iv": "...",
      "authTag": "...",
      "trustQuotient": 0.4533,
      "intent": "preference",
      "similarity": 0.89
    }
  ],
  "recall_mode": "semantic",
  "count": 1
}

AES-256-GCM Encryption by Default

Content is AES-256-GCM encrypted server-side at rest, with a per-operation random IV and GCM authentication tag validated on every read. Embeddings are generated from content before encryption for semantic search (pgvector, 1536 dims, HNSW index).