RobotMem

v1.0.0Knowledge & Memorystable

Enables robots to store, retrieve, and consolidate episodic experiences including physical parameters, trajectories, and outcomes. It supports hybrid vector search with structured filtering and spatial sorting to help robotic agents learn from past s

agent-memoryai-agentsai-memoryclaude-codeepisodic-memory
Share:
24
Stars
0
Downloads
0
Weekly
0/5

What is RobotMem?

RobotMem is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to enables robots to store, retrieve, and consolidate episodic experiences including physical parameters, trajectories, and outcomes. it supports hybrid vector search with structured filtering and spatia...

Enables robots to store, retrieve, and consolidate episodic experiences including physical parameters, trajectories, and outcomes. It supports hybrid vector search with structured filtering and spatial sorting to help robotic agents learn from past s

This server falls under the Knowledge & Memory category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • Enables robots to store, retrieve, and consolidate episodic

Use Cases

Robotic episodic memory
Experience replay learning
Hybrid vector search
robotmem

Maintainer

LicenseApache 2.0
Languagepython
Versionv1.0.0
UpdatedApr 30, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx robotmem

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use RobotMem

RobotMem is a specialized MCP server that gives robotic AI agents persistent episodic memory backed by a local SQLite database with vector search capabilities. It allows robots and autonomous agents to store physical experiences (parameters, trajectories, outcomes), retrieve relevant past experiences using hybrid BM25 plus vector search with spatial sorting, and learn from repeated tasks over time. RobotMem is particularly useful for manipulation tasks, locomotion experiments, and any scenario where an agent should improve from its history of successes and failures.

Prerequisites

  • Python 3.9+ installed
  • pip for installing the robotmem package
  • An MCP-compatible client such as Claude Desktop or Claude Code
  • Sufficient disk space in ~/.robotmem/ for the SQLite memory database (no GPU required)
1

Install RobotMem

Install the robotmem Python package from PyPI. This installs the server and CLI tools.

pip install robotmem
2

Verify installation and launch the web UI

Confirm the install succeeded and optionally open the web UI to browse stored memories.

robotmem --version
robotmem web
3

Add to Claude Desktop config

Configure robotmem as an MCP server in Claude Desktop. The database is stored at ~/.robotmem/memory.db by default.

{
  "mcpServers": {
    "robotmem": {
      "command": "python",
      "args": ["-m", "robotmem"],
      "env": {}
    }
  }
}
4

Start a memory session

Begin an episode by calling start_session with context metadata about the robot and current task.

start_session(context='{"robot_id": "arm-01", "task": "pick-and-place"}')
5

Store and retrieve experiences

Use learn() to record insights and save_perception() to store sensor data. Use recall() with hybrid BM25+vector search to retrieve relevant past experiences.

learn(insight="grip_force=12.5N yields highest grasp success rate",
      context='{"params": {"grip_force": {"value": 12.5, "unit": "N"}}, "task": {"success": true}}')

memories = recall(query="grip force parameters",
                  context_filter='{"task.success": true}',
                  spatial_sort='{"field": "spatial.position", "target": [1.3, 0.7, 0.42]}')
6

End the session

Call end_session to close the episode, which automatically triggers memory consolidation and proactive recall of relevant past experiences.

end_session(session_id=session["session_id"])

RobotMem Examples

Client configuration

Claude Desktop configuration for the RobotMem MCP server using the installed Python module.

{
  "mcpServers": {
    "robotmem": {
      "command": "python",
      "args": ["-m", "robotmem"],
      "env": {}
    }
  }
}

Prompts to try

Example prompts for robotic memory workflows using RobotMem.

- "Store what we learned from today's grasping experiments: the optimal grip force was 12.5N for cylindrical objects"
- "Recall all past experiences where the robot arm succeeded at pick-and-place near coordinates [1.3, 0.7, 0.4]"
- "What grip strategies have worked best for delicate objects based on past sessions?"
- "Forget the incorrect memory about using 20N force for foam objects"
- "Start a new session for the locomotion experiment on terrain type: gravel"

Troubleshooting RobotMem

Vector search returns no results even with relevant stored memories

Vector embeddings are generated using FastEmbed ONNX models on first run, which downloads model files. Ensure internet access is available on first use and allow time for the model download to complete.

Database file grows very large over many sessions

Use the forget() tool to remove incorrect or outdated memories. You can also use context_filter queries to scope recall operations and use the robotmem web UI to inspect and prune the memory database at ~/.robotmem/memory.db.

CJK characters in memory queries return no results

RobotMem includes jieba tokenization for CJK full-text search. Ensure you are on version 0.1.0+ where CJK support is included. If issues persist, also provide an English translation in your recall query.

Frequently Asked Questions about RobotMem

What is RobotMem?

RobotMem is a Model Context Protocol (MCP) server that enables robots to store, retrieve, and consolidate episodic experiences including physical parameters, trajectories, and outcomes. it supports hybrid vector search with structured filtering and spatial sorting to help robotic agents learn from past s It connects AI assistants to external tools and data sources through a standardized interface.

How do I install RobotMem?

Follow the installation instructions on the RobotMem GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with RobotMem?

RobotMem works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is RobotMem free to use?

Yes, RobotMem is open source and available under the Apache 2.0 license. You can use it freely in both personal and commercial projects.

Browse More Knowledge & Memory MCP Servers

Explore all knowledge & memory servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "robotmem": { "command": "npx", "args": ["-y", "robotmem"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use RobotMem?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides