MCP Server

What is MCP?

Model Context Protocol (MCP) lets AI clients like Claude Desktop and Cursor connect to external tools. Mengram's MCP server gives these clients persistent memory.

Setup with Claude Desktop

Add to your Claude Desktop config (claude_desktop_config.json):

{{
  "mcpServers": {{
    "mengram": {{
      "command": "uvx",
      "args": ["mengram-ai"],
      "env": {{
        "MENGRAM_API_KEY": "om-your-key"
      }}
    }}
  }}
}}

Setup with Cursor

Add to Cursor's MCP settings:

{{
  "mcpServers": {{
    "mengram": {{
      "command": "uvx",
      "args": ["mengram-ai"],
      "env": {{
        "MENGRAM_API_KEY": "om-your-key"
      }}
    }}
  }}
}}

Available tools

The MCP server exposes 21 tools:

Tool	Description
remember	Save knowledge from conversation to memory
remember_text	Save knowledge from plain text
recall	Semantic search through memory
search	Structured search with scores and facts
search_all	Unified search across all 3 memory types
timeline	Search memory by time range
vault_stats	Memory statistics
run_agents	Run memory agents (curator, connector, digest)
get_insights	AI-generated insights and patterns
list_procedures	List learned workflows/procedures
procedure_feedback	Record success/failure for a procedure
procedure_history	Version history of a procedure
get_entity	Get details of a specific entity
delete_entity	Delete an entity and all its data
list_episodes	List or search episodic memories
get_graph	Get the knowledge graph
get_triggers	List smart triggers and reminders
get_feed	Activity feed — recent memory changes
archive_fact	Archive a specific fact on an entity
merge_entities	Merge two entities into one
reflect	Trigger AI reflection on memories

HTTP transport

For remote/cloud MCP clients, Mengram also supports SSE transport:

SSE endpoint: https://mengram.io/mcp/sse
Messages: https://mengram.io/mcp/messages/