Skip to main content
AllMCPs
BrowseBestCategoriesStackCompareToolsGuidesBlog
Log in Submit MCP

Stay in the loop

Get new MCP servers and top picks in your inbox.

AllMCPs

The open directory for discovering and installing Model Context Protocol servers.

AllMCPs on GitHub (opens in a new tab)
Launched onTiny Startupstinystartups.com
Explore
  • Browse servers
  • Best MCP servers
  • Categories
  • MCP clients
  • Agent prompts
  • Stack Builder
  • Compare servers
  • Random discovery New
  • Submit a server
  • Pricing & Boost Boost
Learn
  • Guides hub
  • What is MCP?
  • Install guide
  • Build an MCP server
  • Deploy an MCP server
  • Security guide
  • Troubleshooting
  • MCP for SEO & AEO
  • Protocol versioning
  • Transports: stdio vs HTTP
  • State of MCP (stats)
  • Blog & updates
Tools
  • All developer tools
  • Config generator
  • Config validator
  • Config auditor
  • MCP playground
  • Token calculator
  • OpenAPI → MCP
  • Badge generator
For agents
  • REST API docs
  • Trust & traffic Live
  • Remote MCP server SSE ↗ (opens in a new tab)
  • llms.txt ↗ (opens in a new tab)
  • Catalog JSON ↗ (opens in a new tab)
Company
  • About
  • Advertise Sponsor
  • Contact
  • GitHub ↗ (opens in a new tab)
  • Terms
  • Privacy
AllMCPs VerifiedAllMCPs VerifiedFeatured on Nick LaunchesFeatured on Nick LaunchesLaunch Llama NewsletterLaunch Llama NewsletterVerified DR - allmcps.comVerified DR - allmcps.comFeatured on SaaSGrowFeatured on SaaSGrowFeatured on Twelve ToolsFeatured on Twelve ToolsFeatured on Saaspa.geFeatured on Saaspa.geFeatured on Findly.toolsFeatured on Findly.toolsFeatured on Startup FameFeatured on Startup FameFeatured on LaunchKiwiFeatured on LaunchKiwiFeatured on ScrollLaunchFeatured on ScrollLaunchFeatured on DailyPingsFeatured on DailyPingsFazier badgeFazier badgeFeatured on NewTool.siteFeatured on NewTool.siteFeatured on saasfame.comFeatured on saasfame.comDR Checker - Domain RatingDR Checker - Domain RatingListed on Turbo0Listed on Turbo0Launched on LaunchBoard - Product Launch PlatformLaunched on LaunchBoard - Product Launch PlatformList on SimilarlabsList on Similarlabshttps://codetrendy.comhttps://codetrendy.comListed on DevTool.ioFeatured on BuildlistFeatured on BuildlistLaunched on Tiny StartupsFeatured on ShowMeBestAIFeatured on ShowMeBestAIFind us on LaunchZoneFind us on LaunchZoneAllMCPs VerifiedAllMCPs VerifiedFeatured on Nick LaunchesFeatured on Nick LaunchesLaunch Llama NewsletterLaunch Llama NewsletterVerified DR - allmcps.comVerified DR - allmcps.comFeatured on SaaSGrowFeatured on SaaSGrowFeatured on Twelve ToolsFeatured on Twelve ToolsFeatured on Saaspa.geFeatured on Saaspa.geFeatured on Findly.toolsFeatured on Findly.toolsFeatured on Startup FameFeatured on Startup FameFeatured on LaunchKiwiFeatured on LaunchKiwiFeatured on ScrollLaunchFeatured on ScrollLaunchFeatured on DailyPingsFeatured on DailyPingsFazier badgeFazier badgeFeatured on NewTool.siteFeatured on NewTool.siteFeatured on saasfame.comFeatured on saasfame.comDR Checker - Domain RatingDR Checker - Domain RatingListed on Turbo0Listed on Turbo0Launched on LaunchBoard - Product Launch PlatformLaunched on LaunchBoard - Product Launch PlatformList on SimilarlabsList on Similarlabshttps://codetrendy.comhttps://codetrendy.comListed on DevTool.ioFeatured on BuildlistFeatured on BuildlistLaunched on Tiny StartupsFeatured on ShowMeBestAIFeatured on ShowMeBestAIFind us on LaunchZoneFind us on LaunchZone
© 2026 Jackalope Digital LLC. All rights reserved.
  1. Home
  2. 💻 Developer Tools
  3. Task Checkpoint
T
Health: Not checked yetWe have not completed a health check for this listing yet.No health check has run yet.

Task Checkpoint

User RatingsBe the first to rate and review this MCP server! Enrichment pendingWe haven’t run our AI enrichment pass on this listing yet, so the overview, use cases, and FAQ below may be sparse or missing. We work through the catalog over time — check back soon.
View Repository

MCP server that checkpoints long agent tasks: roll back to any step, resume in a new session.

Quick Install

Automated & IDE Setup

Copy the AI prompt to install this server into Claude Code, Cursor, or another agent — or use 1-click editor setup below.

One-click editor setup isn’t available for this listing yet — we don’t have a confirmed install command, and we’d rather show nothing than point your editor at the wrong package or host. Follow the project’s own setup instructions, linked above.

Manual Client & Custom JSON ConfigExpand JSON ▾
No confirmed setup config for this listing yet. We only publish a config block when the install details come from the project itself — its README, its docs, or a verified owner. We haven’t found those for Task Checkpoint, and we’d rather show nothing than a guess you’d paste into your client. Follow the project’s own setup instructions for the current steps.
Install Directory Badge Claim listing Alternatives💻 More in Developer Tools

Documentation Overview

Task Checkpoint MCP

PyPI npm License: MIT Python 3.10+ Dependencies: 0 MCP server

给长任务做"做一步、存一步"的存档:中断后新会话能接着做,做错了能退回任意一步。

一个 stdio MCP 服务器,只用 Python 标准库,不装任何第三方包。

快速开始 | 工具 | 客户端配置 | 使用限制 | English

快速开始

一、装

bash
pipx install task-checkpoint-mcp

二、在客户端里接上(Claude Code、Codex、Cursor 等都是同一段配置)

config.json
{
  "mcpServers": {
    "task-checkpoint": { "command": "tc-mcp", "args": [] }
  }
}

三、把 SKILL.md 装进技能目录,不装的话 Agent 不会主动存档

bash
tc-mcp --install-skill ~/.claude/skills/task-checkpoint

SKILL.md 随包一起装(wheel 里的包数据),不需要仓库、也不需要联网。想看内容用 tc-mcp --print-skill,想知道它在哪用 tc-mcp --skill-path。

(技能目录按客户端而定,见下面"把 SKILL.md 装进 Agent 的技能目录"。)

四、交给 Agent 用。 一个任务就四步:

Code
tc_init    开任务
tc_save    每完成一步存一次(带 conclusion 和 next)
tc_resume  新会话开头先调这个接上进度
tc_restore 退回去(先 apply:false 预览,再 apply:true)

五、不用记工具名,直接说人话。

用 task-checkpoint 开个任务,目标是把 user 模块拆成 service 和 repository 两层。每完成一步存一次档,带结论和下一步。

它会调 tc_init 建任务,之后每完成一步调一次 tc_save。中途换会话、或者隔天接着做,你说一句"接着上次的做",它调 tc_resume 就把进度拿回来了。

("每完成一步就存档"这条约束的可靠来源是 SKILL.md —— 装进技能目录之后它是硬规则;不装,就只能靠你每次提醒。)

中途被打断、隔天换个模型重开,新会话调一次 tc_resume 拿到的是这个:

Code
任务:重构认证模块
目标:把 session 改成 JWT
当前步骤:3
下一步:把三个视图函数切到新入口,删旧函数
未落档:src/views/auth.py, src/views/user.py

handoff 字段就是这段文本,可以直接粘给新会话。它同时告诉你哪几个文件改了但还没落档。

为什么不直接用 git stash 或随手 commit 一下

做法缺什么
git stash一次性的,不能命名、不能跨会话交接,也没有"这一步在做什么、结论是什么"
随手一个 commit会污染真实历史;半成品未必允许 commit;还得你记得先 commit
手写进度笔记和文件状态脱钩,回退时要自己对着时间线拼
本工具文件变化后台自动记(变更层),步骤语义由模型声明(任务层);只额外加 refs/checkpoints/...,不动你的 git 历史

值得装:多步、可能被中断、需要能退回的编码或文档任务。 不值得装:一次能做完、不需要回退也不需要交接的改动。

装之前先知道

  • 需要 Python 3.10 或更新,只用标准库。
  • 不碰你的 git 历史:不 commit、不 reset、不 clean、不动已有分支和 HEAD,只额外加 refs/checkpoints/... 引用。
  • 不是 git 仓库也能用,功能一样,只是少了那层 git 引用。
  • 不保存聊天记录,只保存工作区文件和模型主动声明的步骤说明。
  • 存档默认写在工作区的 .checkpoints/;指定外部 store 时,工作区仅留一个不含文件内容的 .task-checkpoint-store.json 路径指针。
  • 敏感文件识别是路径黑名单,盖不全。 别把凭据放在被扫路径里 —— 详见"使用限制"。

安装

方式一:装成命令(推荐)

bash
pipx install task-checkpoint-mcp
# 或
pip install task-checkpoint-mcp

也可以直接从 git 装,跟着仓库走、不走 PyPI 的版本号:

bash
pipx install "git+https://github.com/qddfxp/task-checkpoint-mcp"

装完在客户端里用 tc-mcp 启动。包里有两个等价的入口点:tc-mcp 和 task-checkpoint-mcp(后者是给 uvx 这类按包名找命令的 runner 用的):

config.json
{
  "mcpServers": {
    "task-checkpoint": { "command": "tc-mcp", "args": [] }
  }
}

方式二:直接用源码(不用安装)

bash
git clone https://github.com/qddfxp/task-checkpoint-mcp
config.json
{
  "mcpServers": {
    "task-checkpoint": {
      "command": "/absolute/path/to/python",
      "args": ["/absolute/path/to/task-checkpoint-mcp/scripts/tc_mcp.py"]
    }
  }
}

command 要写解释器的绝对路径,不要写 python —— 客户端不一定能解析到你要的那个。装过之后也可以直接 python -m tc_mcp。

方式三:用 npx 启动

npm 上有一个同名的 task-checkpoint-mcp,但它是启动器,不是服务器本体:

Terminal
npx -y task-checkpoint-mcp

它只做一件事:找到本机装了 tc_mcp 的 Python 解释器,然后把 stdio 透传给 python -m tc_mcp。所以本机仍然要先有 Python 包(方式一或方式二)。没有的话它会打印安装命令、以退出码 1 结束,而不是丢一堆 traceback 给你。

适合客户端只认 npm 式启动命令的情况:

config.json
{
  "mcpServers": {
    "task-checkpoint": { "command": "npx", "args": ["-y", "task-checkpoint-mcp"] }
  }
}

npx -y task-checkpoint-mcp --version / --help 是启动器自己的开关,不需要本机有 Python。

配置文件放哪

上面那段 mcpServers JSON 要写进客户端的配置文件。各客户端的位置和顶层键:

客户端配置文件顶层键
Claude DesktopWindows %APPDATA%\Claude\claude_desktop_config.json;macOS ~/Library/Application Support/Claude/claude_desktop_config.jsonmcpServers
Claude Code项目根目录 .mcp.jsonmcpServers
Cursor全局 ~/.cursor/mcp.json;项目内 .cursor/mcp.json 优先mcpServers
Windsurf~/.codeium/windsurf/mcp_config.jsonmcpServers
ZCode~/.zcode/cli/config.jsonmcpServers
VS Code / Copilot项目内 .vscode/mcp.json;用户级 %APPDATA%\Code\User\mcp.jsonservers
Codex CLI~/.codex/config.tomlTOML 的 [mcp_servers],不是 JSON

两个容易踩的:

VS Code 的顶层键是 servers,不是 mcpServers。 直接抄上面的 JSON 不会生效,得改键名:

config.json
{
  "servers": {
    "task-checkpoint": { "command": "tc-mcp", "args": [] }
  }
}

Codex 用 TOML。 同样一段配置要写成:

toml
[mcp_servers.task-checkpoint]
command = "tc-mcp"
args = []

其余客户端以各自的文档为准。

怎么用

最重要的一条:文件变化由后台线程自动记录,但"这一步在做什么、结论是什么、下一步干什么"只有模型主动调 tc_save 才会留下。

所以要让 Agent 每完成一步就存一步 —— 这份约束写在 SKILL.md 里,得把它装进 Agent 的技能目录(见下面"把 SKILL.md 装进 Agent 的技能目录")。不装它,文件回退点照常产生,但没人知道当初要干什么、下一步该干什么。

四条铁律:

  1. 开长任务前先 tc_init,一个任务一个名。已有活动任务会被挂起而不是关闭,可以 tc_switch 切回去。
  2. 每完成一步就 tc_save,并且必须填 conclusion 和 next。不填,新会话就只看得到一堆文件路径。
  3. 新会话、或中断后继续时,第一件事是 tc_resume,不要从零猜进度。
  4. 回退前先预览:先 apply: false 看清单,确认了再 apply: true。

一步的粒度:能独立说清"做完了、结论是 X"的单元。太细(每个文件一次)会淹没有效信息,太粗(整个任务一次)就失去了回退的意义。

工具

10 个。第一个参数都是 root,指工作区目录(通常是当前项目的绝对路径)。

工具什么时候用必填
tc_init长任务开工。已有活动任务会被挂起而不是关闭root, name
tc_save每完成一步root, title
tc_resume新会话开头、接手别人中断的工作(只读)root
tc_show想看有哪些步骤 / 变更记录root
tc_restore退回某一步或某条变更记录。先预览再执行root,加 index 或 drift_id
tc_capture想立刻记一次文件变化(不等后台线程)root
tc_switch回到之前挂起的任务root, task_id
tc_export导出交接包给另一个目录 / 另一台机器root, to
tc_import导入别人的交接包,在这里接续任务root, package
tc_compress存档太大了,回收旧变更层空间root

tc_save 的完整参数:

  • title(必填)——这一步一句话标题,写"做了什么",不写"改了什么"
  • conclusion——这一步的结论。最有价值的字段
  • next——下一步要干什么。接续工作的关键
  • description——为什么这么做(决策理由,事后没人记得)
  • verified——已验证项列表(跑了什么测试、确认了什么事实)
  • open_questions——还没解决、留给后面的问题
  • close: true——收尾时用。关闭后不能再存档,但还能读

回退操作本身也会被记成一条变更记录,返回里的 recovered 就是它的 drift_id —— 退错了可以再用它退回来。

返回值里几个值得看的字段

  • tc_resume.handoff —— 一段现成的交接文本,可以直接粘给新会话;drift_paths 是工作区里还没落档的改动;health.errors 是存档自身的问题
  • tc_save.idempotent —— 同一步(标题、文件摘要、结论、下一步、已验证项都一样)重复存,不会产生重复步骤
  • tc_save.active_task_changed —— 传了别人的 task_id 时活动任务会被切过去(原任务变 suspended),这个字段就是在提示这件事
  • tc_restore.recovered —— 回退本身也会被记成一条变更记录,这就是它的 drift_id
  • tc_restore.plan.unrestorable / skipped_paths —— 目标里没能还原的文件;预览和执行给的是同一份清单
  • tc_capture.waiting / partial —— 还在静默期,以及中间那条不完整的记录
  • tc_init.exclude —— 回读你这次设进去的排除模式
  • tc_export.path —— 交接包目录,里面有一个 HANDOFF.md,是给接手方的提示词

把 SKILL.md 装进 Agent 的技能目录

SKILL.md 是这个项目的另一半功能,不是可选文档:MCP 服务器负责存取,SKILL.md 负责让模型每完成一步真的去调 tc_save。

复制到客户端扫描的技能目录,目录名即技能名:

客户端放这里
Claude Code~/.claude/skills/task-checkpoint/SKILL.md
Codex CLI~/.codex/skills/task-checkpoint/SKILL.md
ZCode~/.zcode/skills/task-checkpoint/SKILL.md
跨客户端共用~/.agents/skills/task-checkpoint/SKILL.md

~/.agents/skills/ 是多个客户端共用的技能目录;客户端专属目录优先于它 —— 同名技能以客户端专属目录里的那份为准。OpenCode 等其它客户端放到它自己文档里的技能目录即可,规则不变。

bash
tc-mcp --install-skill ~/.claude/skills/task-checkpoint

SKILL.md 是随包一起安装的(wheel 里的包数据),所以 pipx install 的用户不需要仓库,也不需要联网。目标已存在时会拒绝覆盖 —— 确实要覆盖加 --force;只想看内容用 tc-mcp --print-skill。

克隆了仓库的话,cp SKILL.md <技能目录>/ 效果一样。

Windows 路径形如 C:\Users\<你>\.claude\skills\task-checkpoint\SKILL.md。各客户端的技能目录可能不同,以它自己的文档为准;要求只有一条:文件落在技能目录下的 <技能名>/SKILL.md。重启客户端后生效。

自检(可选)

装完最快的一条,确认入口点活着、拿到版本号:

bash
tc-mcp --version        # 版本号
tc-mcp --help           # 所有开关

下面几条都要先克隆仓库 —— pip 装出来的包里没有 tests/ 和 tools/。

在克隆下来的仓库里跑一遍回归测试:

bash
python -m unittest discover -s tests -p "test_*.py" -v

最后一行是 OK 就对了。整套测试(100 多个用例)跑一次的时间完全看环境:

跑在哪实测
CI 的 ubuntu runner4.7 秒
CI 的 windows runner39.8 秒
本机 Windows(E 盘)156~340 秒

同一套代码差了七十倍。开销集中在 git 子进程(refs/checkpoints/ 相关用例占大头)和静默期等待,对磁盘和杀毒的实时扫描极其敏感 —— 别把秒数当承诺,也别拿一个数字去对比另一台机器。

全部只用标准库,不需要 pytest;在 Python 3.10 上也能跑,只有读 pyproject.toml 的那几条会因为 tomllib 被跳过(3.11+ 全跑)。

仓库里还能跑真机端到端演练(起真 MCP 子进程、真杀进程、建约 190 MB 的重工作区、逐项核对"不动你 git"的承诺):

bash
python tools/drill.py

它跑完默认把场地删掉;想看场地就 TC_DRILL_KEEP=1 python tools/drill.py(演练有未通过项时也会自动保留,方便排查)。

确认 MCP 服务器能起来 —— 会回一行带 serverInfo 的 JSON:

bash
echo '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2025-06-18"}}' | python scripts/tc_mcp.py

从 pip 装的版本没有 scripts/,把上面那条改成 ... | tc-mcp(或 python -m tc_mcp)就行。

环境变量

变量默认说明
TC_STORE<工作区>/.checkpoints存档目录
TC_WATCHon设成 off 关掉后台自动记录
TC_WATCH_INTERVAL15后台检查间隔(秒)
TC_DRIFT_MAX_WAIT60静默期上限(秒)
TC_GITon设成 off 就完全不建 refs/checkpoints/... 引用
TC_GIT_VERIFYon建引用前后对比 git 指纹,自证"没动过用户 git"。设成 off 跳过自证,每次 tc_save 少几趟 git 子进程;跳过后返回的 git.unchanged 是 null
TC_SHA_REUSEoffmtime+size 没变的文件不再重读、直接复用索引里的 sha。风险见"使用限制"

使用限制

敏感文件:黑名单,盖不全

敏感文件不存内容、连哈希都不存,只能看出 mtime 和 size 变了没有。判定按路径匹配,所以两头都会漏:

  • 漏判:名字无辜的凭据抓不到。config/database.yaml、docker-compose.yml、config.json 会被当普通文件明文存进 payload/。别把凭据放在被扫路径里;避不开时用 tc_init(exclude=[...]) 把整个目录排掉。
  • 误拦:被误判为敏感的源码永远拿不到内容副本,回退时会静默跳过(只出现在 plan.unrestorable 里)。名单分两层就是为压低误拦。

拦的范围:

Read the full README →View source on GitHub →

Related MCP Servers

View all in Developer Tools View all alternatives
  • O
    Openapi MCP Server

    Connect any HTTP/REST API server using an Open API spec (v3)

    💻 Developer Tools3 views
    Compare vs Openapi MCP Server →
  • C
    Claude Task Master

    AI-powered task management system for AI-driven development. Features PRD parsing, task expansion, multi-provider support (Claude, OpenAI, Gemini, Perplexity, xAI), and selective tool loading for optimized context usage.

    💻 Developer Tools8 views
    Compare vs Claude Task Master →
  • M
    MCP Server Docker

    Integrate with Docker to manage containers, images, volumes, and networks.

    💻 Developer Tools3 views
    Compare vs MCP Server Docker →
  • A
    Andrea9293 MCP

    Local-first document management and semantic search for AI coding agents

    💻 Developer Tools2 views
    Compare vs Andrea9293 MCP →

Reviews

No reviews yet — be the first to share how this listing worked for you.

Frequently Asked Questions about Task Checkpoint

We don't have a confirmed install command for Task Checkpoint yet, so we don't publish a generated one — a guessed package name would point at the wrong package or none at all. Follow the project's own README or setup instructions (https://github.com/qddfxp/task-checkpoint-mcp) for the current steps.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewTask Checkpoint AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/task-checkpoint?style=directory)](https://allmcps.com/mcp/task-checkpoint)
HTML Embed
<a href="https://allmcps.com/mcp/task-checkpoint"><img src="https://allmcps.com/api/badge/task-checkpoint?style=directory" alt="Task Checkpoint on AllMCPs" /></a>

Technical Specs & Signals

Category💻Developer Tools
More technical detailsExpand ▾
Last updatedSep 28, 2026
Views0
Unique ViewsTotal visits recorded for this listing page on AllMCPs.
Installs0
Installs & Copy ActionsTotal times users copied install commands or configuration snippets for this server.
27Quality signal: Emerging · 27/100How this signal is calculated ▾
Server availabilityNot measured

Not scored for repo-hosted servers — we can't reach the running server, only its GitHub page. Hosted MCP endpoints are health-checked live.

Verified ownership8/20
Documentation & tools11/30
Adoption & activity1/15
Community engagement0/10

A guidance signal from public completeness & health data — not a user rating. New listings start lower and rise as they add docs, get verified, and grow adoption. Signals we can't observe for a listing are skipped, not counted against it.

★ Featured
A

AllMCPs Server

The official MCP server for AllMCPs.com - submit and manage tools directly from your AI. The open directory for MCP servers. Connect Claude, Cursor, Windsurf, and AI agents to databases, tools, files, and APIs. Explore 10,000+ servers. AllMCPs is the premier, open directory for discovering, evaluating, and installing Model Context Protocol (MCP) servers to equip AI agents and LLMs with real-world superpowers.

Explore Server →

Own this project?

This directory is pre-filled from public sources. Claim via GitHub README, site badge, or DNS TXT to unlock edit access and the Official badge and attach your website — proof is checked automatically, then reviewed by our team.

Free dofollow backlink: add your website and place the AllMCPs badge on it — no claim needed. We detect it automatically and keep it verified as long as the badge stays live.

Claim & get free dofollow

Share & Embed

Add our SVG badge (dark/light directory styles) or embeddable widget to your site.

Explore more

More in 💻 Developer Tools →Best MCP servers for Developers →Alternatives to Task Checkpoint →Install in Claude DesktopInstall in CursorInstall in VS CodeSetup guides for all 13 MCP clients