原型 · 预览版:收录官方注册中心里的全部服务器,每 30 分钟同步一次。评价和账户功能将在后续阶段推出。
PopNx MCP为使用 AI 的人准备的 MCP 服务器 · 用大白话讲清楚

Kythene

作者:kythene版本 1.0.2i★ –i在云端运行
注册名称com.kythene/kythene

安装

概览

作者提供的说明(英文原文)

Your AI's work, reviewed by your team, remembered - shared artifacts with review, plus team memory.

PopNx MCP 的三大支柱

这三项永远不会合并成一个分数:意见、测量和人气是不同的东西。

它能帮你做什么

我们还没有为它写通俗的介绍。下面是我们从注册中心读取到的事实。

  • 连接方式: 在云端运行
  • 传输方式: Streamable HTTP
  • 分类: 搜索与 AI 记忆

安装

选择你使用的 AI 应用,然后复制设置。

安装之前

你需要

你用的是哪个应用?

在上方选择你的应用。我们会记住它,下次按钮会显示“安装到 …”。

适用于网页版、桌面版和手机版 Claude:

  1. 打开 Settings → Connectors
  2. 点击 Add custom connector
  3. 粘贴下面的地址,然后点击 Add
  4. 如果要求登录,请登录作者的服务
SettingsConnectors1+ Add2https://…3
https://kythene.com/mcp/kythene

在终端中运行这条命令

claude mcp add --transport http kythene https://kythene.com/mcp/kythene

在 Cursor 中打开

或者自己复制设置

粘贴到项目中的 .cursor/mcp.json,或 ~/.cursor/mcp.json

{
  "mcpServers": {
    "kythene": {
      "url": "https://kythene.com/mcp/kythene"
    }
  }
}

浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。

在 VS Code 中打开

或者自己复制设置

粘贴到项目中的 .vscode/mcp.json

{
  "servers": {
    "kythene": {
      "type": "http",
      "url": "https://kythene.com/mcp/kythene"
    }
  }
}

浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。

ChatGPT 只能连接互联网上的服务器:

  1. 打开网页版 ChatGPT → Settings → Connectors (Apps)
  2. 点击 Create 或 Add(部分套餐需要先在 Advanced 中开启 Developer mode)
  3. 粘贴下面的地址,起个名字,然后保存
  4. 如果要求登录,请登录作者的服务
SettingsConnectors1+ Create2https://…3
https://kythene.com/mcp/kythene

菜单名称可能因套餐和版本不同而略有差异。

你的选择通过 Cookie 保存在这台设备上,无需账户。

这些设置是根据作者在注册中心的条目生成的。如果 README 的写法不同,请以 README 为准。

它能做什么

我们还没有读取这个服务器的工具列表;我们会轮流检查各个端点。

基础检查i

我们检查的是任何人都能看到的信息;我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月6日

  • ✓ 通过

    作者与仓库一致

    注册名称已通过域名 kythene.com 验证(注册中心会核实域名所有权)

  • … 尚未检查

    仍在维护

    我们还没有从 GitHub 读取这个仓库

  • … 尚未检查

    有许可证

    我们还没有从 GitHub 读取这个仓库

  • … 尚未检查

    端点有响应

    尚未检查;我们会轮流检查各个端点

  • … 尚未检查

    使用 OAuth 登录

    尚未检查;我们会轮流检查各个端点

  • … 尚未检查

    工具描述中没有隐藏指令

    尚未检查;我们会在检查端点时读取工具列表

与“搜索与 AI 记忆”中的其他服务器比较i

  • 星标数多于此分类中其他 7246 个服务器的 0%
  • 发布到注册中心的时间比此分类中其他 7246 个服务器的 30% 更新
  • 基础检查结果等于或优于此分类中其他 7246 个服务器的 100%

与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时,星标数才计入。

常见问题

Kythene 安全吗?

它在 1 项适用的基础检查中通过了 1 项。我们没有阅读代码。在添加密钥或令牌之前,请先看看这个服务器要求哪些权限,并使用只具备所需权限的密钥。

它需要哪些密钥或代码?

注册中心没有列出必需的密钥,不过有些服务器会在首次使用时要求你登录。

它在我的电脑上运行还是在云端?

它在作者的云端运行。通过互联网连接,无需安装,但你发送的内容会经过作者的服务器。

它还在维护吗?

GitHub 仓库最后更新于–,最新版本(1.0.2)于3个月前进入注册中心。

哪些应用可以使用它?

支持 MCP 的应用,例如 Claude Desktop、Claude Code、Cursor 和 VS Code。它通过 URL 连接,所以接受远程 MCP 服务器的应用也可以使用,例如开启开发者模式的 ChatGPT。

有什么可以替代它?

同一分类中有:Context7, Codebase Memory, Exa

替代选择

同一分类中其他作者的服务器,按星标数排序。

如何看卡片:★ GitHub 星标i三大支柱:评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i

Context7

作者:upstash

Up-to-date docs for AI agents

电脑或云端★ 64,224MIT今天更新
安装

Codebase Memory

作者:DeusData

Codebase knowledge graph for AI agents — 162 languages, sub-ms queries, 99% fewer tokens.

在你的电脑上运行★ 48,148MIT20天前更新
安装

Exa

作者:exa-labs

Connect AI agents to Exa for web search, content fetching, and multi-step research.

在云端运行★ 5,213MIT上个月更新
安装

作者的 README

打开 README

按原文从作者的仓库复制。PopNx MCP 没有审查或背书其内容。HTML、图片和脚本已被移除。链接会离开本站。

Kythene

Everything an AI-native team needs to work together.

kythene.com · Docs · Pricing · Self-hosting


Kythene is the all-in-one for an AI-native team working together with their AI: the work they share and review, a memory and wiki they all recall, the facts and decisions they've settled, the asks that pass between people and agents, and the toolkits and integrations each person's AI is given. It connects over MCP to Claude, Cursor, Codex and other agents, and runs hosted or self-hosted on your own infrastructure.

Connect in one paste

Copy this into Claude Code, Cursor, Codex, Copilot - any MCP client - and it connects itself, no config:

You are my AI assistant. Connect yourself to Kythene by fetching

https://kythene.com/setup-prompt.md and doing exactly what it says.

MCP endpoint

https://kythene.com/mcp/kythene

A standard remote, streamable-HTTP MCP server. Auth is OAuth 2.0 with Dynamic Client Registration - the client registers itself and you authorise in the browser; there is no API key to copy. Self-host installs serve the same endpoint at their own domain.

Tools

Kythene exposes 73 MCP tools. This list is generated from the running server, so it is exactly what the current release ships:

  • `add_contact` - Add a contact: display_name, and identities - each a channel (email, slack, phone, site: ) and an id on it.
  • `answer_question` - Settle an open question with what the source said: the same record becomes a fact, keeping the question.
  • `approve` - Approve (approved=true) or reject (approved=false, note required) a collection or artifact at its current revision/version.
  • `ask_for_help` - Send this workspace's team a request for help, a bug report or a feature request, with what you were doing attached.
  • `ask_question` - Record an open QUESTION: something we need from a named source and don't have.
  • `assign_action` - Say who owes an action.
  • `brief` - Open a session with ONE call instead of four.
  • `catch_up` - See what changed since your agent last looked, AND what has been put on your plate - call this at the START of a session to open caught up.
  • `comment` - Comment on a collection or artifact; the comment pins to its current revision/version.
  • `correct_record` - Fix a record in place - for when we wrote it down wrong, not when the source changed it (that is replace_record).
  • `create_collection` - Create a collection from one or more artifacts - make work known to the workspace (kythe it).
  • `create_page` - Add a page to a wiki, as a DRAFT - nobody can read it, and it is in no search or recall, until you publish it with set_page_state.
  • `create_share_code` - Mint a share code for a tag (a label or a project) - a private link for someone outside the workspace.
  • `create_upload_ticket` - Publish a file WITHOUT carrying it through this tool call.
  • `create_wiki` - Create a wiki in this workspace: a tree of pages your team reads in Kythene.
  • `deprecate` - Mark a memory stale by id so recall stops surfacing it (agents stop applying it), while it stays retrievable for audit - prefer this over forget when knowledge…
  • `edit_collection` - Edit a collection without republishing (which would create a new collection and abandon its comments and history).
  • `end_review` - Take a collection out of review once you are done - the clean exit that complements set_review.
  • `forget` - Permanently remove a memory by id - a fact, question or decision too (only who recorded it or a workspace admin may; deleting one that replaced another puts…
  • `get_action` - Read one action, and with include_messages its conversation: internal notes, replies (drafts until a person approves them) and what came in from outside,…
  • `get_artifact` - Get an artifact's metadata and version history; set include_content to fetch the bytes of a version (0 = latest).
  • `get_collection` - Get a collection with its member artifacts and tags.
  • `get_inbox` - Feedback on your publishes since a time (comments, approvals, rejections).
  • `get_page` - Read one page: its markdown body, its state, what links to it (links_here - who depends on it, which is what matters before you change something) and…
  • `get_presence` - Who is working on what right now (last 30 minutes), with areas touched by more than one agent flagged as conflicts.
  • `get_record` - Get one fact, open question or decision by id, with its source, date, how, status, area, evidence, any correction and the records it replaced or was replaced…
  • `get_skill` - Read one skill.
  • `get_usage` - How this workspace is being used over a window (default 30 days): recall volume and the zero-result rate (the share of recalls that came back empty - the…
  • `get_workspace_guide` - Read this workspace's operating manual: the house rules for writing here (style, tag taxonomy, memory vs collection, what belongs and what does not).
  • `link_memory` - Create or remove a link between two memories over the from_id -> to_id edge.
  • `list_actions` - Read the queue.
  • `list_collections` - List the collections visible in your workspace, newest first.
  • `list_contacts` - List the workspace's contacts - people who ask or are asked for things but never sign in (customers, callers) - with how each is reached.
  • `list_facts` - List this workspace's facts, open questions and decisions, newest first - check here before you ask a client something or assume something we may already know.
  • `list_pages` - Walk a wiki's tree: every page you can see, in tree order - each page followed by its own children, siblings in the order set with reorder_pages - with its…
  • `list_pending` - Your personal inbox across ALL your workspaces (#126): items addressed to YOU - approvals/rejections and comments on your work, block feedback, and memories…
  • `list_projects` - List the projects (project-kind tags) in the workspace - the valid project values for recall, remember and create_collection.
  • `list_readers` - Which agents read a collection or artifact (lineage).
  • `list_share_codes` - List share codes.
  • `list_share_views` - When a collection was opened through a share link, newest first: each view's time, which link (share_code_label names who it was made for), the collection's…
  • `list_skills` - List this workspace's skills.
  • `list_spaces` - Your member workspaces - the valid share targets, and the values space takes.
  • `list_wikis` - List the wikis in a workspace, with what each is for, how many pages it holds, how many of those are published, and when anything in it last went live - enough…
  • `message_action` - Write on an action.
  • `move_page` - Move a page under a different parent, rename its slug, or both.
  • `pin_page` - Pin a page to its workspace (pinned true) or unpin it (pinned false).
  • `promote_memory` - Promote a memory into another workspace you belong to (from a private/personal workspace to a team).
  • `publish_skill` - Publish a skill for this workspace - a SKILL.md and its files - so the AI of every member whose toolkits hold it gets it over the workspace's connector on its…
  • `push_version` - Push a new version of an existing artifact.
  • `raise_action` - Ask somebody for something - the queue flows BOTH ways, so this is how you ask a person as well as how they ask you.
  • `recall` - Recall the most relevant context, with full content, in one call.
  • `record_decision` - Record a DECISION we took - an assumption we are working to is a decision, not a fact.
  • `record_fact` - Record a FACT: what a named source - a client, a document, their site - told us, never our own guess.
  • `remember` - Store a memory (markdown body).
  • `reorder_pages` - Set the order of the pages directly under one parent - the order the tree shows them in, so a reader meets them in the order they should be read.
  • `replace_record` - Replace a settled fact or an in-effect decision with a new one of the same kind - when the source changed what they said, or we decided afresh.
  • `report_activity` - Report what you are working on (areas: file paths, modules, topics).
  • `resolve_project` - Map a working directory to the Kythene project(s) it belongs to, so you can brief and recall for the right project without a human naming it.
  • `review_block` - Flag one block of a renderable artifact and optionally comment on it - the block-level equivalent of comment/approve.
  • `revoke_share_code` - Revoke a share code by id (from list_share_codes).
  • `set_action_priority` - Make an action "low", "normal" or "high" priority.
  • `set_action_state` - Move an action along by naming the target state: "in_progress" means you have started, which is yours to say as the assignee; "pending" means not now, or…
  • `set_collection_state` - Move a collection through its lifecycle by naming the target state: "archived" archives a live collection (a reversible retirement that hides it from…
  • `set_page_state` - Move a page through its lifecycle by naming the target state: "published" makes the newest version the one readers see, and puts it into search and recall…
  • `set_review` - Turn the approval-review flow on or off for an existing collection (requested=true to request review, false to cancel).
  • `set_skill_state` - Move a skill through its lifecycle by naming the target state: "published" publishes the draft as the next version, for the members whose toolkits hold it -…
  • `set_verification` - Change how a published wiki page or a memory is kept current, by id: every puts it on a schedule (verify every 30, 90, 180 or 365 days - never verified, it's…
  • `set_wiki_state` - Archive, restore or delete a whole wiki by naming the target state: "archived" archives a live wiki - it leaves list_wikis and every reader, share link,…
  • `set_workspace_guide` - Replace this workspace's operating manual with body (markdown), and return the updated guide.
  • `share_collection` - Map a collection into another workspace you belong to (e.g. a shared client workspace), or remove that mapping.
  • `share_wiki` - Share a whole wiki with someone outside the workspace: mint a private link that admits its holder to the wiki's published pages - bar any section restricted to…
  • `update_page` - Replace a page's body with body, in FULL - this is not an append.
  • `verify` - Say a published wiki page or a memory is still right, by id - call it once you have checked it against the live system (the code, the config, the running…

Clients

Claude (Code, desktop, web), Cursor, Codex, GitHub Copilot, and any other MCP-capable assistant.

Pricing

  • Free solo (hosted) - Free, one signed-in user
  • Team (hosted) - $15 per user / month, or $150 per user / year
  • Free self-host - Free, one signed-in user
  • Self-host Team - $150 per user / year, billed annually · from 5 users

Enterprise (hosted or self-host) is contact-us. Full pricing: [www.kythene.com/pricing](https://www.kythene.com/pricing).


Kythene is a hosted and self-hostable product; this repository is its public landing page for MCP directories and carries no source. Licensed under BSD-2-Clause.

问 PopNx