Skip to main content

PowerMem MCP Server - Model Context Protocol server for PowerMem memory management

Project description

PowerMem MCP Server

PowerMem MCP Server - Model Context Protocol server for PowerMem memory management.

English | 简体中文

Prerequisites

Before using the PowerMem MCP Server, please ensure:

  1. PowerMem is installed: The server requires PowerMem to be installed. You can install it via:

    pip install powermem
    
  2. Configuration file exists: Create a .env file in your working directory or project root with PowerMem configuration. The server will automatically search for .env files in:

    • Current working directory's .env
    • Project root directory's .env
    • examples/configs/.env

    You can copy the .env.example file as a template:

    cp powermem_mcp/.env.example .env
    

    Then edit the .env file and configure the following key settings:

    • Database Provider: Choose from sqlite, oceanbase, or postgres
    • LLM Provider: Choose from qwen, openai, or mock
    • Embedding Provider: Choose from qwen, openai, or mock
    • API Keys: Set your LLM and Embedding API keys
    • Agent Configuration: Configure memory management settings
    • Intelligent Memory: Enable/configure Ebbinghaus forgetting curve settings
    • Performance: Adjust batch sizes, cache settings, and search parameters
    • Security: Configure encryption and access control
    • Logging: Set logging levels and file paths

    The .env.example file contains detailed comments and example configurations for different use cases.

For PowerMem installation and configuration, please refer to:

PowerMem Documentation

Startup

Support for multiple types of MCP

You can start PowerMem MCP with different protocols using the following commands:

uvx powermem-mcp sse # sse mode, default port 8000 (recommended)
uvx powermem-mcp stdio # stdio mode
uvx powermem-mcp sse 8001 # sse mode, specify port 8001
uvx powermem-mcp streamable-http # streamable-http mode, default port 8000
uvx powermem-mcp streamable-http 8001 # streamable-http mode, specify port 8001

Usage

Use with MCP Client, must use a client that supports Prompts, such as: Claude Desktop. Before entering a request, you need to manually select the required Prompt, then enter the request.

Claude Desktop config example:

{
  "mcpServers": {
    "powermem": {
      "url": "http://{host}:8000/mcp"
    }
  }
}

Available Tools

The PowerMem MCP Server provides the following memory management tools:

  • add_memory: Add new memory to storage. Supports string, message dict, or message list format. Can use intelligent mode for automatic inference.
  • search_memories: Search memories by query text with optional filters, limit, and similarity threshold.
  • get_memory_by_id: Get a specific memory by its ID.
  • update_memory: Update the content and metadata of an existing memory.
  • delete_memory: Delete a specific memory by its ID.
  • delete_all_memories: Batch delete memories by user_id, agent_id, or run_id.
  • list_memories: List all memories with pagination support (limit and offset) and optional filters.

Community

When you need help, you can find developers and other community partners at https://github.com/oceanbase/powermem.

When you discover project defects, please create a new issue on the issues page.

License

For more information, see LICENSE.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

powermem_mcp-0.1.0.tar.gz (10.8 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

powermem_mcp-0.1.0-py3-none-any.whl (10.8 kB view details)

Uploaded Python 3

File details

Details for the file powermem_mcp-0.1.0.tar.gz.

File metadata

  • Download URL: powermem_mcp-0.1.0.tar.gz
  • Upload date:
  • Size: 10.8 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for powermem_mcp-0.1.0.tar.gz
Algorithm Hash digest
SHA256 7645fb15b80d43986f787a3e4044e02ca69499f17b929fd838113cd34a987082
MD5 f6d0907cb3a8d0bec48ce24e60246aa5
BLAKE2b-256 354b4729a5db136c969b1bdab9184f30de40c3bd2cdf03a5f9156552938500ed

See more details on using hashes here.

Provenance

The following attestation bundles were made for powermem_mcp-0.1.0.tar.gz:

Publisher: publish.yaml on oceanbase/awesome-oceanbase-mcp

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file powermem_mcp-0.1.0-py3-none-any.whl.

File metadata

  • Download URL: powermem_mcp-0.1.0-py3-none-any.whl
  • Upload date:
  • Size: 10.8 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.7

File hashes

Hashes for powermem_mcp-0.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 c21e68d32bcb311ae379e91da7289a4de231b520016e0302229f0f24c6319a18
MD5 75acafdb8580cc79be6e11277064b639
BLAKE2b-256 93258f89d2c56f2def1de20734e9c5e0a7ceb8ac34db4a4028f17f2843a73628

See more details on using hashes here.

Provenance

The following attestation bundles were made for powermem_mcp-0.1.0-py3-none-any.whl:

Publisher: publish.yaml on oceanbase/awesome-oceanbase-mcp

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page