MCP Server Setup
Two ways to connect: Streamable HTTP (zero install, remote) or the mcp-remote stdio bridge for clients that require it. Both reach the same 13-tool surface.
On this page
Prerequisites
- ▸A Synapse Layer account at forge.synapselayer.org
- ▸A Connect Token (
sk_connect_*) from Dashboard → Connect - ▸For the stdio bridge: Node.js 18+ — the
mcp-remotebridge runs vianpx
Remote (Streamable HTTP)
No local installation required. Point your MCP client to the hosted endpoint:
json
{
"mcpServers": {
"synapse-layer": {
"url": "https://forge.synapselayer.org/api/mcp",
"transport": "streamable-http",
"headers": {
"x-connect-token": "sk_connect_your_token"
}
}
}
}Note
Streamable HTTP is the recommended transport for most use cases. It requires no local Python installation and connects directly to the Forge infrastructure.
stdio bridge (mcp-remote)
Some hosts (Claude Desktop among them) cannot send custom auth headers on a remote connection. For those, use the mcp-remote bridge — a local stdio process that proxies to the remote endpoint:
json
{
"mcpServers": {
"synapse-layer": {
"command": "npx",
"args": [
"mcp-remote",
"https://forge.synapselayer.org/api/mcp",
"--header",
"x-connect-token: sk_connect_YOUR_TOKEN"
]
}
}
}The first run downloads mcp-remote from npm. There is no local server to install — the bridge talks to the same hosted endpoint.
Environment Variables
| Variable | Required | Description |
|---|---|---|
| SYNAPSE_TOKEN | No | Read by the Python SDK when token= is omitted (Synapse()). Connect token (sk_connect_*) |
Verify Connection
After configuring your MCP client, ask the agent to run health_check. A successful response confirms the connection:
json
{
"status": "healthy",
"version": "2.4.x",
"engine": "synapse-memory",
"user_id": "your-tenant-id"
}