Connect any MCP-compatible client to Pensyve Cloud with a single command. No binary to install, no local storage to manage — just your API key.
The remote MCP server uses Streamable HTTP transport. Your API key authenticates every request and scopes memory to your account namespace.
Prerequisites
- Sign up for a Pensyve Cloud account
- Create an API key at Settings → API Keys
- Set the key in your environment:
export PENSYVE_API_KEY=psk_live_...Connect Your Client
Claude Code
claude mcp add pensyve \
--transport http \
--header "Authorization: Bearer $PENSYVE_API_KEY" \
https://api.pensyve.com/mcpClaude Desktop
Add to your config file (~/Library/Application Support/Claude/claude_desktop_config.json on macOS):
{
"mcpServers": {
"pensyve": {
"url": "https://api.pensyve.com/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_KEY"
}
}
}
}Cursor
Add to .cursor/mcp.json:
{
"mcpServers": {
"pensyve": {
"url": "https://api.pensyve.com/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_KEY"
}
}
}
}VS Code (Copilot Chat)
Add to .vscode/mcp.json:
{
"servers": {
"pensyve": {
"url": "https://api.pensyve.com/mcp",
"headers": {
"Authorization": "Bearer YOUR_API_KEY"
}
}
}
}Available Tools
The remote MCP server exposes 8 tools:
| Tool | Metered | Description |
|---|---|---|
pensyve_recall | Yes | Semantic + BM25 memory search |
pensyve_remember | Yes | Store a fact as semantic memory |
pensyve_episode_start | Yes | Begin an interaction episode |
pensyve_episode_end | Yes | Close an episode |
pensyve_forget | Yes | Delete an entity's memories |
pensyve_inspect | Yes | List memories for an entity |
pensyve_status | Free | Namespace info, memory counts, health |
pensyve_account | Free | Plan, usage, and quota |
Every metered tool call counts as one operation against your plan's usage pool. See Pricing for details.
For full tool parameter reference, see MCP Tools.
Verify Connection
Once connected, ask your agent to check the connection:
"What's my Pensyve status?"
The agent will call pensyve_status and return your namespace, memory counts, and health status.
Remote vs Local
| Remote (Cloud) | Local | |
|---|---|---|
| Setup | API key only | Install binary + ONNX models |
| Storage | Managed Postgres | Local SQLite |
| Embeddings | Cloud GTE (768d) | Local ONNX (GTE/MiniLM) |
| Billing | Metered per operation | Free, unlimited |
| Requires internet | Yes | No |
| Best for | Teams, production agents | Offline dev, privacy-sensitive |
Both modes expose the same 6 core tools with identical parameters and response shapes. Agents don't need to change prompts or workflows when switching between local and remote.