genesys-memory — Claude Code MCP server: features and setup
This MCP server gives an AI agent persistent memory linked by cause and effect. Use it when context needs to carry across tasks.
claude mcp add --env OPENAI_API_KEY=<OPENAI_API_KEY> --env ANTHROPIC_API_KEY=<ANTHROPIC_API_KEY> --env GENESYS_EMBEDDER=<GENESYS_EMBEDDER> --transport stdio genesys-memory -- uvx genesys-memoryWhat genesys-memory does
genesys-memory stores context across tasks and connects facts through causal relationships. It exposes memory to an agent over MCP and is described as having 13 tools, including support for explaining stored memories.
The Claude Code setup runs uvx genesys-memory. Its install command passes OPENAI_API_KEY and ANTHROPIC_API_KEY; an alternate command also sets GENESYS_EMBEDDER. Those values need to be configured for the shown setup to run.
What it helps with
- carry project context between agent tasks
- connect stored facts through causal relationships
- retrieve information saved in earlier work
- explain why a particular memory is relevant
Example prompts for the agent
- Remember that we chose PostgreSQL for this project and why.
- Recall why we rejected the previous approach.
- Connect this decision to the reason behind it.
How to install genesys-memory
Platform CLI
claude mcp add --env OPENAI_API_KEY=<OPENAI_API_KEY> --env ANTHROPIC_API_KEY=<ANTHROPIC_API_KEY> --env GENESYS_EMBEDDER=<GENESYS_EMBEDDER> --transport stdio genesys-memory -- uvx genesys-memoryReplace <…> with your own values. uvx runs the command named after the package; if the README names another one, use uvx --from genesys-memory <command>
Platform CLI
codex mcp add genesys-memory --env OPENAI_API_KEY=<OPENAI_API_KEY> --env ANTHROPIC_API_KEY=<ANTHROPIC_API_KEY> --env GENESYS_EMBEDDER=<GENESYS_EMBEDDER> -- uvx genesys-memoryReplace <…> with your own values. uvx runs the command named after the package; if the README names another one, use uvx --from genesys-memory <command>
Config file
{
"mcpServers": {
"genesys-memory": {
"command": "uvx",
"args": [
"genesys-memory"
],
"env": {
"OPENAI_API_KEY": "<OPENAI_API_KEY>",
"ANTHROPIC_API_KEY": "<ANTHROPIC_API_KEY>",
"GENESYS_EMBEDDER": "<GENESYS_EMBEDDER>"
}
}
}
}Replace <…> with your own values. uvx runs the command named after the package; if the README names another one, use uvx --from genesys-memory <command>
Config file
{
"servers": {
"genesys-memory": {
"type": "stdio",
"command": "uvx",
"args": [
"genesys-memory"
],
"env": {
"OPENAI_API_KEY": "<OPENAI_API_KEY>",
"ANTHROPIC_API_KEY": "<ANTHROPIC_API_KEY>",
"GENESYS_EMBEDDER": "<GENESYS_EMBEDDER>"
}
}
}
}Replace <…> with your own values. uvx runs the command named after the package; if the README names another one, use uvx --from genesys-memory <command>
Platform CLI
gemini mcp add -e OPENAI_API_KEY=<OPENAI_API_KEY> -e ANTHROPIC_API_KEY=<ANTHROPIC_API_KEY> -e GENESYS_EMBEDDER=<GENESYS_EMBEDDER> genesys-memory uvx genesys-memoryReplace <…> with your own values. uvx runs the command named after the package; if the README names another one, use uvx --from genesys-memory <command>
Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about genesys-memory
How do I install genesys-memory in Claude Code?
Run claude mcp add genesys-memory -e OPENAI_API_KEY=<OPENAI_API_KEY> -e ANTHROPIC_API_KEY=<ANTHROPIC_API_KEY> -- uvx genesys-memory.
What does genesys-memory do?
It provides an AI agent with persistent memory connected by causal relationships, exposed through MCP. The description lists 13 tools.
Which environment variables does the Claude Code setup use?
The shown command passes OPENAI_API_KEY and ANTHROPIC_API_KEY. An alternate command also includes GENESYS_EMBEDDER.
