CC-MCP

Name: CC-MCP
Author: Beginnersguide138

@Beginnersguide138

代码仓库

关于 CC-MCP

🌟 Revolutionary AI Context Management: Solving LLM Memory Loss in Long Conversations

#task-never-forget-task-memory

基本信息

分类

其他

传输方式

stdio

发布者

Beginnersguide138

提交者

Beginnersguide138

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "cc-mcp": {
      "autoApprove": [
        "process_user_message",
        "start_session",
        "get_debug_info",
        "list_sessions",
        "get_session_stats",
        "export_context",
        "import_context",
        "clear_context",
        "end_session"
      ],
      "disabled": false,
      "timeout": 120,
      "type": "stdio",
      "command": "uv",
      "args": [
        "run",
        "--directory",
        "/path/to/your/cc-mcp",
        "main.py"
      ],
      "env": {
        "CLASSIFIER_API_URL": "https://api.openai.com/v1/chat/completions",
        "CLASSIFIER_API_KEY": "your_openai_api_key_here",
        "CLASSIFIER_MODEL": "gpt-4o-mini"
      }
    }
  }
}

工具

未检测到工具

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is CC-MCP?

CC-MCP is a context management system for MCP clients that solves LLM “intent forgetting” in long conversations. It provides tools to maintain conversational consistency, remember constraints and decisions, and eliminate the need for users to constantly re-state context across dialogue sessions.

How to use CC-MCP?

Install CC-MCP and configure it with your MCP client (requires Node.js 18+). The server runs locally; details for specific commands or configuration keys are not provided in the README.

Key features of CC-MCP

Intelligent context management for long conversations.
Prevents “intent forgetting” in LLM agents.
Maintains user goals, budgets, and timelines across turns.
Eliminates context fragmentation between dialogue sessions.
Provides persistent memory for repeated interactions.
Works with any LLM that supports MCP.

Use cases of CC-MCP

Project planning and design sessions with evolving constraints.
Customer support that remembers user history across tickets.
Creative writing where character details must stay consistent.
Multi-turn decision-making with strict budget or timeline rules.
Any application requiring long-term conversational coherence.

FAQ from CC-MCP

What is CC-MCP and how is it different from other MCP servers?

CC-MCP focuses specifically on context retention and memory management across long conversations, whereas other servers may handle data access or external API calls without addressing LLM forgetting.

What runtime or dependencies does CC-MCP require?

You need Node.js 18+ and an MCP-compatible client. No other external runtime dependencies are mentioned in the README.

Where does CC-MCP store conversation data?

All data remains local; CC-MCP does not send context to third-party servers. Privacy is maintained by keeping memory on the user’s machine.

Does CC-MCP work with any LLM model?

Yes, as long as the LLM is accessed through an MCP client. The server provides context management tools independent of the underlying model.

What transport and authentication does CC-MCP use?

The README does not specify transport mechanisms or authentication details. CC-MCP runs as a local MCP server, so transport is typically standard MCP (stdio or HTTP).