Ainative Zerodb Memory MCP server
{
"mcpServers": {
"ainative-studio-ainative-zerodb-memory": {
"command": "npx",
"args": [
"-y",
"ainative-zerodb-memory-mcp@latest"
],
"env": {
"ZERODB_API_KEY": "<zerodb_api_key>",
"ZERODB_API_URL": "https://...",
"ZERODB_PROJECT_ID": "<zerodb_project_id>"
}
}
}
}{
"mcpServers": {
"ainative-studio-ainative-zerodb-memory": {
"command": "npx",
"args": [
"-y",
"ainative-zerodb-memory-mcp@latest"
],
"env": {
"ZERODB_API_KEY": "<zerodb_api_key>",
"ZERODB_API_URL": "https://...",
"ZERODB_PROJECT_ID": "<zerodb_project_id>"
}
}
}
}{
"mcpServers": {
"ainative-studio-ainative-zerodb-memory": {
"command": "npx",
"args": [
"-y",
"ainative-zerodb-memory-mcp@latest"
],
"env": {
"ZERODB_API_KEY": "<zerodb_api_key>",
"ZERODB_API_URL": "https://...",
"ZERODB_PROJECT_ID": "<zerodb_project_id>"
}
}
}
}{
"servers": {
"ainative-studio-ainative-zerodb-memory": {
"command": "npx",
"args": [
"-y",
"ainative-zerodb-memory-mcp@latest"
],
"env": {
"ZERODB_API_KEY": "${input:zerodb_api_key}",
"ZERODB_API_URL": "${input:zerodb_api_url}",
"ZERODB_PROJECT_ID": "${input:zerodb_project_id}"
}
}
},
"inputs": [
{
"id": "zerodb_api_key",
"type": "promptString",
"description": "Zerodb api key",
"password": true
},
{
"id": "zerodb_api_url",
"type": "promptString",
"description": "Zerodb api url",
"password": true
},
{
"id": "zerodb_project_id",
"type": "promptString",
"description": "Zerodb project id",
"password": true
}
]
}{
"mcpServers": {
"ainative-studio-ainative-zerodb-memory": {
"command": "npx",
"args": [
"-y",
"ainative-zerodb-memory-mcp@latest"
],
"env": {
"ZERODB_API_KEY": "<zerodb_api_key>",
"ZERODB_API_URL": "https://...",
"ZERODB_PROJECT_ID": "<zerodb_project_id>"
}
}
}
}file .mcp.json (in your project root)file claude_desktop_config.jsonfile ~/.cursor/mcp.jsonfile .vscode/mcp.jsonfile ~/.codeium/windsurf/mcp_config.jsoninstall npx -y ainative-zerodb-memory-mcpversion 1.2.7
What it does
This server provides persistent memory management for AI agents with 14 tools for storing, searching, and retrieving conversation context, plus write-back actions to Slack, Gmail, Calendar, and GitHub. It's for developers building agents that need cross-session memory with semantic search and automatic context pruning to reduce token consumption.
- Store and retrieve user preferences and conversation history across multiple chat sessions using semantic search
- Automatically manage agent context to stay within LLM token limits while keeping important memories
- Write agent decisions or summaries back to external services like Slack channels, Gmail drafts, calendar events, or GitHub issues
Tools this server exposes
| Tool | Does |
|---|---|
| zerodb_store_memory | Store conversation context with automatic importance scoring and vector embeddings |
| zerodb_search_memory | Search memory semantically by meaning across sessions with similarity scores and importance filters |
| zerodb_get_context | Retrieve full conversation context with automatic pruning to fit token limits and memory decay |
| zerodb_embed_text | Generate vector embeddings for text using BAAI BGE models in three dimensions (384, 768, 1024) |
Before you enable it
An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.