mcp-memory-service MCP — 사용 사례, 설치 & 라이브 데모

왜 쓰나요

핵심 기능

Semantic search with tag filtering over durable memories
Backends: SQLite-vec (local), ChromaDB, Cloudflare, hybrid
5ms local retrieval; background consolidation with decay
REST API (15 endpoints) for use outside MCP clients
OAuth 2.0 + PKCE for browser-based claude.ai integration

라이브 데모

실제 사용 모습

memory-service.replay ▶ 준비됨

0/0

설치

클라이언트 선택

~/Library/Application Support/Claude/claude_desktop_config.json · Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "memory-service": {
      "command": "uvx",
      "args": [
        "mcp-memory-service"
      ],
      "_inferred": true
    }
  }
}

Claude Desktop → Settings → Developer → Edit Config 열기. 저장 후 앱 재시작.

~/.cursor/mcp.json · .cursor/mcp.json

{
  "mcpServers": {
    "memory-service": {
      "command": "uvx",
      "args": [
        "mcp-memory-service"
      ],
      "_inferred": true
    }
  }
}

Cursor는 Claude Desktop과 동일한 mcpServers 스키마 사용. 프로젝트 설정이 전역보다 우선.

VS Code → Cline → MCP Servers → Edit

{
  "mcpServers": {
    "memory-service": {
      "command": "uvx",
      "args": [
        "mcp-memory-service"
      ],
      "_inferred": true
    }
  }
}

Cline 사이드바의 MCP Servers 아이콘 클릭 후 "Edit Configuration" 선택.

~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "memory-service": {
      "command": "uvx",
      "args": [
        "mcp-memory-service"
      ],
      "_inferred": true
    }
  }
}

Claude Desktop과 같은 형식. Windsurf 재시작 후 적용.

~/.continue/config.json

{
  "mcpServers": [
    {
      "name": "memory-service",
      "command": "uvx",
      "args": [
        "mcp-memory-service"
      ]
    }
  ]
}

Continue는 맵이 아닌 서버 오브젝트 배열 사용.

~/.config/zed/settings.json

{
  "context_servers": {
    "memory-service": {
      "command": {
        "path": "uvx",
        "args": [
          "mcp-memory-service"
        ]
      }
    }
  }
}

context_servers에 추가. 저장 시 Zed가 핫 리로드.

claude mcp add memory-service -- uvx mcp-memory-service

한 줄 명령. claude mcp list로 확인, claude mcp remove로 제거.

사용 사례

실전 활용법: mcp-memory-service

How to give Claude stable memory of your preferences and context

👤 Daily power users tired of re-explaining themselves ⏱ ~10 min beginner

언제 쓸까: Every new chat starts from scratch and you want continuity.

사전 조건

uvx installed — brew install uv

흐름

Seed preferences

memory_store: 'I work on iOS + Swift projects. Prefer MVVM with @Observable. Always comment in English.' with tags [preference, coding].✓ 복사됨

→ Stored confirmation
In a future session, retrieve

memory_search: 'my coding style preferences'.✓ 복사됨

→ Preferences returned, agent applies them
Decay cleanup

Show me memories older than 90 days with no recent hits — offer to prune.✓ 복사됨

→ Stale list for review

결과: Claude behaves like it remembers you.

함정

Storing sensitive PII that becomes retrievable forever — Don't — treat memory like a public notebook. Use tags to compartmentalize.

How to share a memory pool across a team via Cloudflare backend

👤 Teams running shared AI workflows ⏱ ~30 min advanced

언제 쓸까: You want every engineer's chat to remember the same architectural decisions.

사전 조건

Cloudflare account with Workers KV or D1 — cloudflare.com, set up as per the Cloudflare backend docs

흐름

Switch backend

Configure MCP_MEMORY_BACKEND=cloudflare with CF creds in env.✓ 복사됨

→ Memory endpoints point to CF
Seed team-shared memory

memory_store: 'Our auth lives in services/auth. Always rotate JWT keys via the rotate-keys Make target.' with tag 'team-arch'.✓ 복사됨

→ Stored
Verify from a second machine

memory_search 'auth service' — should return the same entry.✓ 복사됨

→ Cross-device hit

결과: Shared institutional memory that survives team turnover.

함정

Sensitive data in shared memory leaks cross-team — Run separate backends per team; don't store secrets

How to auto-harvest learnings at the end of a coding session

👤 Solo devs using agents heavily ⏱ ~5 min beginner

언제 쓸까: At the end of a pair-programming chat, before closing.

흐름

Run the harvester

memory_harvest on this conversation — extract durable facts (API keys patterns, project decisions, known gotchas). Ignore chit-chat.✓ 복사됨

→ Structured list of candidate memories
Approve + save

Save items 1, 3, 5 as memories with tag 'project-x'. Drop the rest.✓ 복사됨

→ Count saved

결과: Low-friction capture of what you learned, nothing more.

조합

다른 MCP와 조합해 10배 효율

memory-service + contextplus

Use Context+ for live repo state, memory-service for durable cross-session facts

Save this architectural decision to memory-service and link it to the relevant files via Context+ memory graph.✓ 복사됨

memory-service + github

After each merged PR, harvest learnings into memory

Summarize the last 10 merged PRs in org/repo, extract repeat mistakes, and store them as memories with tag 'code-review-lessons'.✓ 복사됨

도구

이 MCP가 노출하는 것

도구	입력	언제 호출	비용
memory_store	content: str, tags?: str[]	Save a durable fact	1 embedding
memory_search	query: str, tags?: str[], limit?: int	Retrieve relevant memories	1 vector search (local)
memory_harvest	transcript: str	End-of-session capture	1 LLM call
memory_store_session	session_id, messages	Ingest a whole chat	N embeddings

비용 및 제한

운영 비용

API 쿼터: SQLite-vec: unlimited local. ChromaDB: local. Cloudflare: Workers/KV/D1 free tier is generous.
호출당 토큰: Store: ~100 tokens for content. Search: ~50 tokens per result.
금액: Free (open source). Cloudflare backend: free tier usually enough.
팁: Start with SQLite-vec. Switch to hybrid only when you need multi-device sync.

보안

권한, 시크릿, 파급범위

최소 스코프: Local filesystem for SQLite mode Cloudflare API token for CF mode

자격 증명 저장: CF creds via env vars

데이터 외부 송신: Local (SQLite) or your Cloudflare account (CF)

절대 부여 금지: Don't let untrusted prompts call memory_store — poisoning attacks

Memory is long-lived by default — deleting a memory doesn't erase it from LLM context that already saw it.
OAuth 2.0 + PKCE remote mode exposes an HTTP surface; keep it behind auth.

문제 해결

자주 발생하는 오류와 해결

ModuleNotFoundError: sqlite_vec

pip/uv install with the sqlite-vec extra: uvx 'mcp-memory-service[sqlite]'

확인: python -c 'import sqlite_vec'

Cloudflare backend 403

Token lacks Workers/KV permissions. Create a new CF API token with the specific scopes.

Memories returned aren't relevant

Re-embed with a better model — defaults to local MiniLM; upgrade to nomic-embed via env var.

memory_harvest outputs duplicates

Enable the dedup config; harvester will skip items with cosine similarity > 0.95 to existing.

대안

mcp-memory-service 다른 것과 비교

대안	언제 쓰나	단점/장점
mem0	You want a hosted memory SaaS	Paid beyond free tier; vendor-locked
contextplus memory graph	You want memory tightly integrated with code understanding	Scoped to a repo, not general-purpose

mcp-memory-service

왜 쓰나요

핵심 기능

라이브 데모

실제 사용 모습

설치

클라이언트 선택

사용 사례

실전 활용법: mcp-memory-service

How to give Claude stable memory of your preferences and context

사전 조건

흐름

함정

How to share a memory pool across a team via Cloudflare backend

사전 조건

흐름

함정

How to auto-harvest learnings at the end of a coding session

흐름

조합

다른 MCP와 조합해 10배 효율

도구

이 MCP가 노출하는 것

비용 및 제한

운영 비용

보안

권한, 시크릿, 파급범위

문제 해결

자주 발생하는 오류와 해결

대안

mcp-memory-service 다른 것과 비교

더 보기

리소스