Cogmemai Mcp

Name: Cogmemai Mcp
Author: hifriendbot

@hifriendbot

代码仓库

关于 Cogmemai Mcp

Persistent cognitive memory for Claude Code. Cloud-first MCP server with

#memory #mcp #ai-memory #semantic-search #claude-code #cursor #developer-tools

基本信息

分类

其他

传输方式

stdio

发布者

hifriendbot

提交者

hifriendbot

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "cogmemai": {
      "command": "cogmemai-mcp",
      "env": {
        "COGMEMAI_API_KEY": "your_key_here"
      }
    }
  }
}

工具

Store a fact explicitly (architecture decision, preference, etc.)

Search memories using natural language (semantic search)

Ai extracts facts from a conversation exchange automatically

Load top memories at session start (with smart ranking and compact mode)

Browse memories with filters (paginated)

Update a memory's content, importance, or scope

Permanently delete a memory

Check your usage stats and tier info

Export all memories as JSON for backup or transfer

Bulk import memories from a JSON array

Feed in a document (README, API docs) to auto-extract memories

Save a summary of what was accomplished in this session

View all tags in use across your memories

Connect related memories with named relationships

Explore the knowledge graph around a memory

View edit history of a memory

Memory health dashboard with insights

Promote a project memory to global scope

概览

What is Cogmemai Mcp?

Cogmemai Mcp is a cognitive memory server for Claude Code and other AI coding assistants. It solves the problem of losing context between sessions by persisting architecture decisions, coding preferences, bug fixes, and project knowledge permanently. It is built for developers who want their AI assistant to remember everything without re-explaining.

How to use Cogmemai Mcp?

Run npx cogmemai-mcp setup for an automatic configuration wizard that verifies your API key, configures the MCP server, and installs automatic context recovery. Alternatively, manually add it to .mcp.json or use claude mcp add with your COGMEMAI_API_KEY. It works with Claude Code, Cursor, Windsurf, Cline, and Continue. The server exposes 18 tools that the AI assistant uses automatically, including save_memory, recall_memories, and extract_memories.

Key features of Cogmemai Mcp

Semantic search by meaning, not keywords
AI-powered automatic extraction of important facts
Smart deduplication of conflicting memories
Privacy controls that detect secrets before storage
Document ingestion from READMEs and docs
Project scoping with global and per-repo memories
Automatic compaction recovery for Claude Code
Token-efficient compact context loading
Zero local databases or vector stores

Use cases of Cogmemai Mcp

Remembering your tech stack and architecture decisions across Claude Code sessions
Preserving coding style preferences and tool choices globally
Tracking bugs, workarounds, and dependency constraints per project
Instantly loading project context from ingested documentation
Maintaining consistent context for long-running or multi-session development tasks

FAQ from Cogmemai Mcp

How do I get an API key?

Get a free API key at hifriendbot.com/developer. The key starts with cm_ and is set as COGMEMAI_API_KEY.

What makes CogmemAi different from local memory solutions?

CogmemAi runs extraction and search server-side. Your MCP server is a thin HTTP client with no local databases, no Docker, no Python, and no vector stores. Setup is zero maintenance.

Does CogmemAi store my source code?

No. CogmemAi stores only extracted facts (short sentences) and never raw source code. API keys are hashed with SHA-256, all traffic is over HTTPS, and no data is used for model training.

What editors and assistants are supported?

Cogmemai Mcp works with Claude Code, Cursor, Windsurf, Cline, and Continue. Automatic setup is recommended for Claude Code; manual JSON configuration is available for the others.

How does compaction recovery work?

When Claude Code compacts your context, CogmemAi automatically preserves your context before compaction and seamlessly restores it afterward, preventing loss of conversation history and context.