Skip to content
MCP server

MemCell

Persistent shared memory for agents that recalls before acting and learns from outcomes

Install
Pricing
Freemium
Platforms
API, CLI, self-hosted
Runs
Locally or Remotely
Version
2.2.1
Listing updated
3 days ago

About MemCell

MemCell is a freemium MCP server for knowledge and memory. It can run locally on your machine or as a hosted remote server.

MemCell scores each memory by how it actually performed. An agent recalls relevant instructions, decisions and facts before it acts, then reports whether the recalled memory worked, failed or was avoided, and the confidence of that memory adjusts. A get_insights tool exposes a calibration curve and governance metrics such as enforced guards and quarantined memories.

The seven MCP tools cover recall, remember, report, feedback, document ingestion, document search and insights. Documents in PDF, DOCX, CSV, TSV, HTML, JSON and Markdown can be indexed into grounded chunks. Memories can be shared across team workspaces and reviewed before going team-wide.

It runs as a remote server at memcell.ai/mcp or locally via npx. SDK adapters exist for the Vercel AI SDK, OpenAI Agents, LangGraph, PydanticAI and CrewAI.

Install

Run in a terminal

claude mcp add --transport http memcell https://memcell.ai/mcp

Add to Settings → Connectors → Add custom connector

https://memcell.ai/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "memcell": {
      "url": "https://memcell.ai/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "memcell": {
      "type": "http",
      "url": "https://memcell.ai/mcp"
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "memcell": {
      "serverUrl": "https://memcell.ai/mcp"
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "memcell": {
      "type": "http",
      "url": "https://memcell.ai/mcp"
    }
  }
}

Tools this server provides (7)

Taken from the project documentation; the running server may differ.

recall
Pre-action retrieval of instructions, decisions, facts and preferences.
remember
Stores durable memories into a workspace or user profile.
report
Reports whether recalled memory worked, failed or was avoided.
feedback
Marks a specific memory as worked or failed with notes.
ingest_document
Indexes documents into chunks and optionally distills memories.
search_documents
Hybrid semantic and full-text search over document chunks.
get_insights
Returns workspace telemetry, a confidence calibration curve and governance metrics.

What it does

Pre-action recall
Fetches relevant instructions, decisions, facts and preferences before an agent acts.
Outcome-based confidence
Reports of worked, failed or avoided adjust each memory's confidence.
Document ingestion
Indexes PDF, DOCX, CSV, TSV, HTML, JSON and Markdown into searchable chunks.
Guard mode
Memories can be enforced as checks at specific moments.
Framework adapters
Drop-in adapters for Vercel AI, OpenAI Agents, LangGraph and CrewAI.
Team workspaces
Shared memory across people and agents, with review before team-wide use.

Good for

  • Make agents follow company rules such as refund sign-off limits
  • Share one memory across agents from different vendors
  • Turn instruction files into retrievable memories
  • Check how reliable the stored knowledge has proven

Pricing

Freemium, from $19 a month

Free tier; Pro $19/mo; Team $19/seat/mo; Enterprise by quote

Prices change. Check memcell.ai before you buy.

MemCell FAQ

What does it cost?

Free covers 10,000 memories and 3 API keys; Pro is $19 a month and Team $19 per seat a month.

What happens at the memory limit?

New memories stop saving with a message; recall keeps working and nothing is deleted.

Can it run offline?

The README describes a local server started with memcell start.

How many tools are there?

Seven.

Is MemCell free?

MemCell has a free plan, with paid plans for more usage or features. Paid plans start at $19 a month.

How do I add MemCell to Claude Code?

Run this command in your terminal: claude mcp add --transport http memcell https://memcell.ai/mcp

What are the best alternatives to MemCell?

The closest alternatives to MemCell are Codebase Memory, XERJ, Claude Code Ultimate Guide, Nautilus Compass Memory and note1.

Alternatives to MemCell

  1. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  2. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  3. MCP server for searching the Claude Code Ultimate Guide from any MCP client

    Runs locally
    • Open source
    • 6.5K GitHub stars
    • AI models and LLM tooling
  4. Local-first long-term memory layer for AI agents, served over MCP

    Not checked yet
    • 5 tools
    • Remote
    • Open source
    • 1.3K GitHub stars
    • Knowledge and memory
  5. note1

    MCP server

    MCP server giving AI assistants search and summaries across your note1 meetings

    • 4 tools
    • Remote
    • Free trial
    • 1 GitHub star
    • Productivity