TeamMemory MCP server
io.github.ysydhc/team-memory·v0.1.2·1 starsTeam experience database - let AI accumulate, retrieve and apply team knowledge across chats.
1 stars45 downloads/wk
Reviews
Write oneNobody has reviewed TeamMemory yet.
If you have run it, two minutes of your experience saves the next person an afternoon.
TeamMemory tools (13, 2 write)
write = sends, deletes, buys or postsRead from the package source without running it. The installed server may list more.
tm_claimMark that you're working on a specific experience or problem.
tm_configReturn current runtime configuration snapshot for diagnostics.
tm_feedbackwrite actionSubmit feedback for an experience.
tm_learnExtract and save an experience from conversation/document text.
tm_notifyNotify the team about a new or updated experience.
tm_saveQuick-save a new experience to the team knowledge base.
tm_save_groupSave a group of experiences (parent with children).
tm_save_typedSave a typed experience with full fields.
tm_searchSearch team experiences by semantic similarity.
tm_solveSolve a problem by searching team experiences with enhanced context.
tm_statusReturn runtime status for troubleshooting.
tm_suggestSuggest relevant experiences based on current working context.
tm_updatewrite actionUpdate an existing experience record.
Public scan report
scanner v0.1.10 · 2026-09-28 · same rubric, same numbers if you re-run it
- Code scan80 source files scanned25/25
- –Live reliabilityno gateway calls yet and no remote to proben/a
- –Tool poisoningtools not inspected (local package is not executed); not countedn/a
- Auth qualitystatic API keys via environment variables6/15
- Maintenancelast push 102 days ago8/15
- Maintainer identityregistry namespace matches repository owner; GitHub account older than a year8/10
How TeamMemory compares
Against every graded Databases MCP server, from our own scans and uptime checks. Updated after each scan run.
- Safety score
- 72/100
- higher than 30% of 802 Databases servers
- Tools
- 13
- the typical Databases server has 8
- Tools that change things
- 15%
- typical: 10%. Fewer write tools means less an agent can break
- GitHub stars
- 1
- more than 45% of 602 Databases servers
- Downloads a week
- 45
- more than 32% of 328 Databases servers with a package
What its tools let an agent do: 11 read · 2 change
Install TeamMemory in Claude Code, Cursor or VS Code
claude mcp add team-memory -- uvx team_memory
What the publisher says
From the TeamMemory repository's README, as published. We do not edit it. Read it on GitHub
TeamMemory
mcp-name: io.github.ysydhc/team-memory
让 AI 拥有团队记忆 — 跨会话积累经验,像资深成员一样理解你的项目。
这是我学 AI 时萌生的一个想法。市面上已有类似产品,但总觉得不太贴合自己的使用习惯。做这个项目,既想通过和 AI 一起写代码来加深对大模型的理解,也希望能按自己的工作流,打磨出真正顺手的功能。
给 Agent / 贡献者:AGENTS.md · docs/README.md · MCP 实现 src/teammemory/server.py · 分层约束见 scripts/harnessimportcheck.py(LAYERMAP)
为什么需要 TeamMemory?
用 Cursor、Claude 等 AI 助手写代码时,往往会遇到三个问题:
TeamMemory 就是冲着这三个问题来的。 通过 MCP 把语义可搜索的经验库接进 AI:遇到问题自动查历史方案,解决后自动提炼并存下来,下次谁遇到同类问题,直接就能命中。既适合 3–10 人的技术团队共享,也适合部署在本地个人使用,配合 Cursor / Claude Desktop。
快速开始(4 条命令 + 1 项配置)
环境:Docker Desktop、Python 3.11+、Make
# 1. 初始化(Docker + 依赖 + 数据库)
make setup
# 2. 设置 API Key(唯一必改项;与 Web 签发的原始密钥同为 64 位十六进制,见 docs/decision/auth-api-key-design.md)
export TEAM_MEMORY_API_KEY=$(openssl rand -hex 32)
echo "API Key: $TEAM_MEMORY_API_KEY"
# 3. 拉取 Embedding 模型(仅首次需要)
ollama pull qwen3-embedding:0.6b
# 4. 启动
make web浏览器访问 http://localhost:9111 ,用上面的 API Key 登录即可。更完整的部署与用户流程见下文 按角色导航 与 快速开始。
CLI 工具 / CLI Tools
除 MCP 外,所有 memory_* 工具也可通过 tm-cli 命令行调用:
# 搜索团队知识 / Search team knowledge
tm-cli recall --query "如何配置"
# 保存经验 / Save experience
tm-cli save --title "Bug fix" --problem "连接超时" --solution "增加重试"
# 获取上下文 / Get context
tm-cli context --file-paths "src/server.py"
# 查看所有命令 / Show all commands
tm-cli --help配置命令 / Configuration Commands
# 显示当前生效配置及值来源 / Show effective config with source annotations
tm-cli config show
# 交互式生成 ~/.config/tm/config.toml / Interactive config init
tm-cli config init
# 一键配置平台 Hook / One-command hook setup
tm-cli setup --platform claude-code
tm-cli setup --platform cursor
tm-cli setup --platform hermes
tm-cli setup --platform all
# 仅检查当前状态 / Check status only
tm-cli setup --platform all --check前提:make dev 启动服务 + TEAMMEMORYAPI_KEY 环境变量已设置(或已在 ~/.config/tm/config.toml 中配置)。
MCP 接入(Cursor / Claude)
本仓库(克隆源码)推荐:不要把 API Key 写进 mcp.json。在仓库根维护 .env(从 example/env.team-memory.example 复制),其中至少设置 TEAMMEMORYAPIKEY;MCP 配置为 bash + scripts/runmcpwithdotenv.sh + cwd= 仓库根。详见 docs/guide/mcp-server.md。Cursor 一般用 .cursor/mcp.json,Claude Code 可用根目录 .mcp.json,两处内容建议保持一致。
项目名零配置:在项目根放置 .tm.toml(见 配置说明),或系统自动从 git 仓库名推断,无需传 --project 或设 TEAMMEMORYPROJECT。
仅 pip install teammemory、无本地仓库目录**时,可在 .cursor/mcp.json 里用本机 Python 与环境变量(数据库与 Key 仍需提供):
{
"mcpServers": {
"team_memory": {
"command": "python3",
"args": ["-m", "team_memory.server"],
"env": {
"TEAM_MEMORY_DB_URL": "postgresql+asyncpg://developer:devpass@localhost:5433/team_memory",
"TEAM_MEMORY_API_KEY": "你的 API Key"
}
}
}
}也可使用客户端配置 ~/.config/tm/config.toml(tm-cli config init 生成),避免在 JSON 中写入密钥。
(MCP 未注册 Resources / Prompts;memory 七工具:memorysave、memoryrecall、memorycontext、memorygetarchive、memoryarchiveupsert、memoryfeedback、memorysubmitresponse。详情见下文 MCP 工具列表(当前) 与 docs/guide/mcp-server.md。)
本机直连数据库时需要配 TEAMMEMORYDBURL(或通过 config);从源码跑且项目里已有 config 的,可不单独设 DBURL。
架构可视化(现状)
Web 内「架构」导航与 /api/v1/architecture/ 已移除(实现见 src/teammemory/web/static/js/pages.js)。若需要代码库图谱,请在本机单独使用 GitNexus(CLI / Bridge 等),与当前 TM Web 无集成。
按角色导航
Shortened. The full README is on GitHub.
Nothing above is checked by us. What we check is on the safety report.
TeamMemory: common questions
- Is TeamMemory MCP server safe?
- Mostly: it is graded B (72/100). Read the TeamMemory safety report
- How do I install TeamMemory?
- It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
- Does TeamMemory need an API key?
- Yes. The registry entry asks for
DATABASE_URL. - Is TeamMemory maintained?
- The last commit was 139 days ago (2026-06-18). The latest release is v0.1.2.
- What can I use instead of TeamMemory?
- Servers from other publishers that do the same job: CogniCore MCP server, LINQ to DB MCP server and open-memex MCP server.
- How does TeamMemory compare with other Databases MCP servers?
- safety score 72/100 (higher than 30% of 802 Databases servers); tools 13 (the typical Databases server has 8); tools that change things 15% (typical: 10%. Fewer write tools means less an agent can break). Full comparison.
Alternatives to TeamMemory
Same job from other publishers: the closest match first, then the best rated.
- CogniCoreMemory, reflection and safety for AI agents. Hosted or local, no vector database.5.0(1)GrowingA
- LINQ to DBInspect database schemas and execute SQL queries across multiple database providers with LINQ to DB.not reviewedEstablishedA
- open-memexLocal-first memory for AI coding agents — Markdown + SQLite FTS5, shared across editors via MCP.not reviewedGrowingB
- EmeraldThe database under the chat: a local SQLite file a model asks for rows, not notes to re-read.not reviewedGrowingA
- CitadelEncrypted-first embedded database with vector search and agent memory, exposed as MCP toolsnot reviewedGrowingB