Power your agent
การติดตั้งจากผู้เผยแพร่ ไม่ใช่เอนด์พอยต์ของ PlugMCP
URL แบบรีโมตหรือคำสั่งบนเครื่องของคุณ? การติดตั้งสองแบบต่างกันอย่างไร
รันบน PlugMCP
A new server you create. It is not this listing's official endpoint. Use the publisher install when you want their URL and their credentials.
เปิดจากผู้เผยแพร่
A Go implementation of the HMLR (Hierarchical Memory Lookup & Routing) memory system, exposed as an MCP (Model Context Protocol) server for integration with Claude Code and other MCP clients.
HMLR replaces traditional vector-only RAG systems with a structured, state-aware memory architecture that achieves across adversarial benchmarks.
Go MCP server for HMLR hierarchical memory system with semantic search
Key Features:
~/.local/share/remember/git clone <your-repo-url>
cd remember-standalone
.env file with your API keys:cp .env.example .env
# Edit .env and add your OPENAI_API_KEY
Example .env:
OPENAI_API_KEY=sk-proj-...
ANTHROPIC_API_KEY=sk-ant-... # Optional, for future features
MEMORY_OPENAI_MODEL=gpt-4o-mini # Optional, defaults to gpt-4o-mini
go build -o bin/hmlr-server ./cmd/server
./bin/hmlr-server
The server will start on stdio and wait for MCP protocol messages.
Required:
OPENAI_API_KEY - Your OpenAI API key for embeddings and LLM featuresOptional:
MEMORY_OPENAI_MODEL - Chat model to use (default: gpt-4o-mini)
gpt-4o-mini, gpt-4o, o1-mini, etc.Model Selection Guide:
gpt-4o-mini: Recommended - Good balance of speed, quality, and cost (~$0.15/1M input tokens)gpt-4o: Highest quality, slower, more expensive (~$2.50/1M input tokens)o1-mini: Advanced reasoning capabilities (~$3/1M input tokens)Add to your Claude Code MCP settings (~/.config/claude-code/mcp_settings.json):
{
"mcpServers": {
"remember": {
"command": "/path/to/remember-standalone/bin/hmlr-server",
"args": [],
"env": {
"OPENAI_API_KEY": "your-key-here",
"MEMORY_OPENAI_MODEL": "gpt-4o-mini"
}
}
}
}
The server exposes 5 MCP tools:
store_conversationStore a conversation turn in HMLR memory.
Input:
{
"message": "What's the capital of France?",
"context": "Optional additional context"
}
Output:
{
"block_id": "block_20251206_143022",
"turn_id": "turn_20251206_143022_abc123",
"routing_scenario": "topic_continuation",
"facts_extracted": 1
}
retrieve_memorySearch for relevant memories.
Input:
{
"query": "What did we discuss about France?",
"max_results": 5
}
Output:
{
"memories": [
{
"block_id": "block_20251206_143022",
"topic_label": "Geography",
"relevance_score": 0.95,
"summary": "Discussion about European capitals",
"turns": [...]
}
],
"facts": [
{
"key": "capital_of_France",
"value": "Paris",
"confidence": 1.0
}
]
}
list_active_topicsList all active conversation topics.
Output:
{
"topics": [
{
"block_id": "block_20251206_143022",
"topic_label": "Geography",
"status": "ACTIVE",
"turn_count": 3,
"created_at": "2025-12-06T14:30:22Z"
}
]
}
get_topic_historyGet full conversation history for a topic.
Input:
{
"block_id": "block_20251206_143022"
}
get_user_profileGet the user profile summary.
Output:
{
"profile": {
"preferences": {},
"topics_of_interest": [],
"last_updated": "2025-12-06T14:30:22Z"
}
}
HMLR Memory System
├── Governor # Smart routing (4 scenarios)
├── ChunkEngine # Hierarchical chunking (turn → paragraph → sentence)
├── Storage # XDG-compliant file + SQLite storage
├── Bridge Blocks # Topic-based conversation organization
└── MCP Server # Stdio transport for Claude Code integration
~/.local/share/remember/
├── bridge_blocks/
│ └── 2025-12-06/
│ ├── block_*.json # Conversation topics
│ └── day_metadata.json
├── facts.db # SQLite fact database
├── embeddings/
│ └── *.json # Vector embeddings
└── user_profile.json # Long-term user profile
# Run all scenario tests (with REAL storage, no mocks!)
go test -v ./.scratch/
# Run specific scenario
go test -v ./.scratch/ -run TestScenario01
remember-standalone/
├── cmd/
│ └── server/ # Main entry point
├── internal/
│ ├── core/ # Governor, ChunkEngine
│ ├── storage/ # Storage implementation
│ ├── models/ # Data structures
│ └── mcp/ # MCP tools and handlers
├── .scratch/ # Scenario tests (not committed)
├── scenarios.jsonl # Documented test scenarios
└── DESIGN.md # Full architecture design
This project follows scenario-driven testing with zero mocks:
# All tests use REAL storage, REAL SQLite, REAL files
.scratch/scenario_01_store_retrieve_test.go # Basic storage
.scratch/scenario_02_routing_test.go # Governor routing
.scratch/scenario_03_chunking_test.go # Hierarchical chunking
See scenarios.jsonl for documented test scenarios.
✅ Phase 1: Foundation
✅ Phase 2: Core Components
✅ Phase 3: Background Agents
✅ Phase 4: MCP Server
✅ Phase 5: Advanced Features
This project uses:
See DESIGN.md for full architecture details.
MIT License - See LICENSE file
Go MCP server for HMLR hierarchical memory system with semantic search
เครื่องมือ
ยังไม่มีรายการเครื่องมือ เซิร์ฟเวอร์แบบรีโมตจะได้รายการหลังการตรวจสอบสถานะครั้งแรก
FAQ
คัดลอกคอนฟิก stdio สำหรับ Cursor, Claude Desktop, Claude Code หรือ VS Code ได้ในหน้านี้ คอนฟิกนี้ใช้การติดตั้งของผู้เผยแพร่ ไม่ใช่พร็อกซีของ PlugMCP
ระเบียนนี้ไม่ได้ระบุตัวแปรสภาพแวดล้อมที่จำเป็น
Memory ทำงานแบบโลคัลผ่าน stdio
ตรวจสอบล่าสุดเมื่อ 2026-10-28 Go MCP server for HMLR hierarchical memory system with semantic search
