Tently

io.github.tently-teamv0.1.1更新于 Oct 5, 2026

Checks your team's architecture decisions before your coding agent writes code.

已验证STDIO仅桌面Developer ToolsKnowledge & Memory

概览

AI 生成的概览

在编码代理写代码之前,向它提供团队的架构决策和本地影响范围分析。

功能
Tently 向编码代理提供两个工具。search_decisions 返回适用于即将修改代码的团队决策,并标明每条是阻断性还是建议性。query_impact_zone 使用 CodeGraph 在本地检出上计算某个符号的影响范围,例如调用方和依赖项。README 指出影响结果是证据而非证明,因为通过接口、动态分派或再导出到达的调用方可能被遗漏。
适用场景
当编码代理在团队已从拉取请求、评审和文档中积累架构规则的代码库中工作时适用,可在拉取请求出现之前检查这些规则。也适合快速查看某个符号的改动可能影响什么。
运行要求
通过 npm 包 @tently/mcp 以 stdio 在本地运行,需要 Node.js 20 或更高版本。需要在 TENTLY_API_KEY 中提供 Tently 工作区 API 密钥,可在 app.tently.dev 的 Connect 页面创建。可选 TENTLY_API_URL(默认 TENTLY_WORKING_DIR。init 命令会修改 .mcp.json 或 .cursor/mcp.json 并添加一个技能;之后需重启代理。
安装前请注意
search_decisions 会把查询文本和你传入的文件路径发送到 Tently API;query_impact_zone 完全在本地运行,源代码不会离开本机。API 密钥是存储于代理配置中的工作区凭据,请妥善保管。init 会修改代理配置文件,但会在编辑前备份文件并打印每处更改。

安装

在 SourceWeft 中

  1. 打开 控制台中的 Tently,将其添加到工作区。
  2. 为需要使用其工具的对话启用该服务。

Desktop only,通过 STDIO。 STDIO 服务会启动本地进程,因此需要 SourceWeft 桌面宿主。

其他 MCP 客户端

参照 仓库 中的启动说明。

README

Tently MCP

[npm] [CI] [License: MIT]

Tently in your coding agent. An MCP server for Claude Code and Cursor that checks your team's decisions (the rules Tently learned from your pull requests, reviews and docs) before the agent writes code.

Coding agents write code fast, but they don't know why your team built things the way it did. Tently remembers, and this server gives your agent that memory while it works, before a pull request exists.

Set up

Create an API key on the Connect page at app.tently.dev, then run this from your repo root:

sh
npx @tently/mcp init --agent claude --api-key <your key>

Or pass the key through the environment:

sh
TENTLY_API_KEY=<your key> npx @tently/mcp init

init adds the tently server to .mcp.json (Claude Code) or .cursor/mcp.json (Cursor), and installs a small skill that tells the agent when to use it. It backs up any file before editing it, keeps your other MCP servers, and prints every change. Restart your agent afterwards.

--agent takes claude, cursor, both or auto (the default, which detects what's in the repo).

Manual config

json
{  "mcpServers": {    "tently": {      "command": "npx",      "args": ["-y", "@tently/mcp", "mcp"],      "env": { "TENTLY_API_KEY": "<your key>" }    }  }}

What the agent gets

  • search_decisions: the decisions that apply to the code it's about to change, from your Tently workspace. Each one says whether breaking it is blocking or advisory.
  • query_impact_zone: the blast radius of a symbol (who calls it, what it depends on), computed locally on your checkout with CodeGraph. Your source code never leaves your machine. The index lives in .codegraph/, which init adds to .gitignore.

Impact results are evidence, not proof: callers reached through interfaces, dynamic dispatch or re-exports can be missed, and the tool says so in its output.

Configuration

VariableMeaning
TENTLY_API_KEYWorkspace API key (required for search_decisions)
TENTLY_API_URLAPI origin, default https://api.tently.dev
TENTLY_WORKING_DIRDirectory query_impact_zone indexes, default the current directory

Privacy

search_decisions sends your query text and the file paths you pass to the Tently API. Nothing else leaves your machine: query_impact_zone runs entirely locally. See tently.dev for how Tently handles data.

Development

sh
npm installnpm testnpm run buildnode dist/cli.js mcp   # start the server on stdio

Requires Node.js 20 or later.

Contributing

Issues and pull requests are welcome. For anything beyond a small fix, please open an issue first so we can agree on the approach.

License

MIT

来源:README.md,提交 bdcbd90

工具

0
工具元数据尚未被收录。

版本历史

1
  1. v0.1.1最新Oct 5, 2026