What can you do with this server?

The provided schema does not match the README: as exposed, this server only offers generic image and URL content analysis, not ZenTao/Life Saver tools. * analyze_image: analyze an image supplied as a string, with an optional hint, returning a string result. * analyze_url: analyze content at a given url, with an optional hint, returning a string result. * It does not expose the README's ZenTao read/search/status/workhours/task/bug operations.

Which integrations are available for this server?

Allows using Google Gemini models for analyzing images and web page content. Allows using OpenAI models (e.g., GPT-4o) for analyzing images and web page content.

How do I use Life Saver MCP?

1. Click on "Deploy Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Life Saver MCP analyze this Zentao bug: https://zentao.example.com/?m=bug&f=view&bugID=1081" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

Life Saver MCP

by kaminof3tt

Overview Schema Source Related Servers Discussions Admin

TypeScript

Remote

life-saver-mcp

禅道（ZenTao）的 MCP 服务器。给任意支持 Model Context Protocol 的客户端用：把禅道上的需求 / 任务 / BUG 连编号一起取出来（正文、备注、操作记录、附件），搜禅道，查工时，也能写回（开始、记工时、完成、解决、关闭、加备注）。

底层取数与写入逻辑在 life-saver-core 里，和 DSH 插件版共用同一份实现——两边的工具描述、参数、排版都来自同一个源，不会各写一套然后慢慢漂移。

六个工具

工具	干什么
`zentao_read`	按编号或链接读一条需求 / 任务 / BUG：正文、备注、操作记录、附件落盘，关联需求一起递归取
`zentao_search`	在需求 / 任务 / BUG 里搜关键词，给出编号与标题
`zentao_status`	体检：两条取数通道配没配齐、能不能连、探到的是哪套禅道
`zentao_workhours`	某账号在任意时间段做了哪些工作、消耗多少工时，以及「有动作但没记工时」的漏项
`zentao_task_op`	写任务：开始 / 记工时 / 完成 / 暂停 / 激活 / 关闭 / 取消 / 备注
`zentao_bug_op`	写 BUG：确认 / 解决 / 关闭 / 备注 / 记工时

写操作会真的写入禅道，没有预览；默认只允许动自己名下的单子（ZENTAO_WRITE_ALLOW_OTHERS=1 才放开）。

Related MCP server: Zentao Bug MCP

两条取数通道

db：直连禅道只读库 + SFTP 取附件。快、能批量、不惊动禅道的审计日志。
http：账号密码登录换 cookie 再走页面。需要账号有对应权限，但和人工看到的一致（页面皮肤差异会自动探测）。
auto（默认）：先 db，失败自动降级到 http，并在结果里说明降级原因。

只配 http 也能用，只配 db 也能用；两个都配最稳（db 挂了还有 http 兜底）。

安装

npm i -g life-saver-mcp

然后在 MCP 客户端里登记。以「命令 + 环境变量」这种配置为例：

{
  "mcpServers": {
    "zentao": {
      "command": "npx",
      "args": ["-y", "life-saver-mcp"],
      "env": {
        "ZENTAO_BASE_URL": "http://zentao.example.com",
        "ZENTAO_ACCOUNT": "your-account",
        "ZENTAO_PASSWORD": "your-password",
        "ZENTAO_DB_HOST": "10.0.0.9",
        "ZENTAO_DB_PORT": "3306",
        "ZENTAO_DB_NAME": "zentao",
        "ZENTAO_DB_USER": "reader",
        "ZENTAO_DB_PASSWORD": "readonly-password"
      }
    }
  }
}

先跑一次体检，确认配置能被解析（不连禅道、不回显密码）：

ZENTAO_BASE_URL=... ZENTAO_ACCOUNT=... life-saver-mcp --check

配置

全部从环境变量读。密码只从环境变量读，不落盘、不进配置文件、不打印。

变量	默认	说明
`ZENTAO_MODE`	`auto`	`auto` / `db` / `http`
`ZENTAO_BASE_URL`	—	禅道根地址，例如 `http://zentao.example.com`（带不带 `/zentao` 后缀都行）
`ZENTAO_ACCOUNT`	—	禅道登录账号
`ZENTAO_SKIN`	`auto`	详情页皮肤：`auto` / `detail-section` / `detail` / `article`；探测总失败时钉死一档
`ZENTAO_HTTP_TIMEOUT_MS`	`60000`	单次 HTTP 超时
`ZENTAO_DB_HOST`	—	只读库地址
`ZENTAO_DB_PORT`	`3306`	只读库端口
`ZENTAO_DB_NAME`	`zentao`	库名
`ZENTAO_DB_USER`	—	只读账号
`ZENTAO_DB_PREFIX`	`zt_`	表前缀
`ZENTAO_DB_CONNECTION_LIMIT`	`4`	连接池上限
`ZENTAO_SFTP_HOST`	—	附件读取用的 SFTP 主机（通常与库同机）
`ZENTAO_SFTP_PORT`	`22`	SFTP 端口
`ZENTAO_SFTP_USER`	—	SFTP 账号
`ZENTAO_SFTP_UPLOAD_BASE`	—	禅道 upload 目录绝对路径，例如 `/var/www/zentao/data/upload`
`ZENTAO_SFTP_PRIVATE_KEY`	—	改用私钥登录时的私钥路径
`ZENTAO_PASSWORD`	—	禅道登录密码
`ZENTAO_DB_PASSWORD`	—	只读库密码
`ZENTAO_SFTP_PASSWORD`	—	SFTP 密码
`ZENTAO_WRITE_ENABLED`	`1`	写能力总开关；置 `0` 即只读
`ZENTAO_WRITE_MODE`	`http`	写走哪条线：`http`（推荐，副作用由禅道自己写全）/ `db` / `auto`
`ZENTAO_WRITE_ALLOW_OTHERS`	`0`	是否允许改不是指派给自己的任务 / BUG
`ZENTAO_DISCOVERY`	`1`	是否自探索部署差异（皮肤、登录算法、库列能力）
`ZENTAO_DISCOVERY_TTL_MS`	`600000`	探测结果缓存时长
`ZENTAO_OUTPUT_DIR`	—	附件落盘目录；留空则放到 `当前目录/zentao-extract/<kind>-<id>`
`ZENTAO_MAX_ATTACHMENTS`	`40`	单个对象最多落盘多少个附件
`ZENTAO_MAX_ATTACHMENT_BYTES`	`20971520`	单个附件大小上限（超过只记录不落盘）
`ZENTAO_MAX_RELATED`	`8`	最多递归取多少个关联对象

也可以用 --config <file> 从 JSON 补一批非密码配置（环境变量优先，密码仍然只认环境变量）。见 life-saver-mcp.example.json。

附件

附件（截图、文档）会落到 ZENTAO_OUTPUT_DIR 或 <当前目录>/zentao-extract/<kind>-<id>/，工具结果里返回绝对路径。MCP 客户端如果支持图片内容块，可以让模型直接看截图；本服务器目前只返回路径与文件名，避免把二进制塞进上下文。

开发

git clone https://github.com/kaminof3tt/life-saver-mcp.git
cd life-saver-mcp
npm i
npm run build        # tsc → lib/
npm test             # 配置解析、tools/list、tools/call 错误路径、真实 stdio 握手

依赖 life-saver-core。本地要联调未发布的核心包时，先把它装进来（只改 node_modules，不动 manifest）：

npm i --no-save ../dsh-life-saver/packages/core

npm run build 之后可以直接 node lib/index.js --check 看配置摘要，或 node lib/index.js 在 stdio 上起服务。

发布

推一个 v* tag，或手动触发 release workflow（workflow_dispatch 带版本号）。workflow 会先确认目标版本的 life-saver-core 已经发布到 npm，再改 manifest 里的版本与依赖、构建、发布，全程不需要跨仓凭证。只在 CI 里发布（NPM_TOKEN 是仓里的 secret），本地直接 npm publish 会绕过版本断言。

另一个 workflow 是 ci：推到 main 与每个 PR 都跑构建、类型检查和测试。在 life-saver-core 上 npm 之前，它会先探一次依赖能不能安装——查不到就只留一条 warning 跳过，不会因为 ETARGET 红一片；等 core 上线后，同一个 workflow 自动开始真正干活。

声明

非官方项目，与禅道官方无关。
写操作直接改生产禅道的数据，请先用只读账号 + ZENTAO_WRITE_ENABLED=0 跑一段时间再放开。
密码只走环境变量；本仓库不含任何真实站点地址、账号或人名。

MIT

Available Tools

2 tools

analyze_imageImage AnalyzerD

Name	Required	Description	Default
`hint`	No
`image`	Yes

Output Schema

Name	Required	Description
`result`	Yes

TDQS

D1/5.0

Behavior1/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Tool has no description.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness1/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Tool has no description.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness1/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Tool has no description.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters1/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Tool has no description.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose1/5

Does the description clearly state what the tool does and how it differs from similar tools?

Tool has no description.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines1/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Tool has no description.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

analyze_urlURL Content AnalyzerD

Name	Required	Description	Default
`url`	Yes
`hint`	No

Output Schema

Name	Required	Description
`result`	Yes

TDQS

D1/5.0

Behavior1/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Tool has no description.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness1/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Tool has no description.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness1/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Tool has no description.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters1/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Tool has no description.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose1/5

Does the description clearly state what the tool does and how it differs from similar tools?

Tool has no description.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines1/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Tool has no description.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

2 tool updatesv0.1.0Jul 6, 2026
- First observedanalyze_image
- First observedanalyze_url

Deploy Server

Pay for results, not tokens.Fund a bounty. Let AI agents compete and collaborate to solve your problems.agentbounties.appAd

TDQS

D1.8/5.0

Scored 2026-07-06 23:05 across 2 tools

Disambiguation5/5

The two tools target different input types (image vs URL), making them clearly distinct despite lacking descriptions.

Naming Consistency5/5

Both tool names follow the consistent 'analyze_' verb_noun pattern.

Tool Count3/5

With only 2 tools, the server feels minimal but may be appropriate for a narrow scope; the name 'Life Saver' suggests a focused utility.

Completeness2/5

The server only offers analysis operations with no CRUD or lifecycle coverage, leaving potential gaps in functionality.

Maintenance

ActivityMaintained

ResponsivenessNo issues

Resources

Related MCP Connectors

anotherai
dev.anotherai
MCP server for building and testing AI agents with multi-model experimentation and insights.
Agent Orchestration Autonomous Agents Testing & QA Tools
2
https://api.anotherai.dev/mcp
Qase Test Management
io.qase
Official MCP server for Qase — manage test cases, runs, suites, defects via AI tools.
Testing & QA Tools Developer Tools
237
https://mcp.qase.io/mcp
SmartBear MCP
com.smartbear
MCP server for AI access to SmartBear tools, including BugSnag, Reflect, Swagger, PactFlow, QTM4J.
Monitoring Testing & QA Tools API Testing
144
https://zephyr.mcp.smartbear.com/mcp
BugEzy A
dev.bugezy
Voice-powered bug reporting with 13 MCP tools. Record bugs by talking; let AI find and fix them.
Speech Processing Testing & QA Tools Coding Agents
142
https://bugezy.dev/mcp

Related MCP Servers

SuperComponents Serverofficial
Developer Tools Design Tools AI & Machine Learning
SuperComponents
A
license
Not graded
quality
D
maintenance
MCP server for AI-powered component generation - analyze designs, generate components, and create implementation instructions.
Updated a year ago (2025-07-13 22:58 UTC)
0
4
MIT
Zentao Bug MCP
Project Management Testing & QA Tools
ceeyang
A
license
A
quality
B
maintenance
MCP server for Zentao that reads bugs, extracts reproduction steps, and allows AI to update bug status after fixing.
Updated 2 months ago (2026-09-07 17:26 UTC)
7
21 npm
6
MIT
Nitpick
Developer Tools Testing & QA Tools
salahashraf3
A
license
Not graded
quality
D
maintenance
A MCP server that enables AI coding agents to consume structured UI feedback via click-to-annotate, supporting issue types and severity levels.
Updated 4 months ago (2026-07-17 05:25 UTC)
8 npm
MIT
lanhu-design-mcp
blantian
A
license
A
quality
C
maintenance
MCP server for macOS that reads Lanhu designs, analyzes UI structures, and exports fine-grained design assets for multiple platforms.
Updated 3 months ago (2026-07-19 15:19 UTC)
8
24 PyPI
5
MIT

life-saver-mcp

六个工具

两条取数通道

安装

配置

附件

开发

发布

声明

Available Tools

Output Schema

TDQS

Output Schema

TDQS

Tool Schema Changelog

TDQS

Maintenance

Resources

Related MCP Connectors

Related MCP Servers

SuperComponents Serverofficial

Zentao Bug MCP

Nitpick

lanhu-design-mcp