Cork

MCP开发工具

提供方 Kevin Vera

开发工具 HTTP · 远程 OAuth 授权 社区 / 第三方 Claude DesktopClaudeCursorVS Code
查看接入方式

Cork is the multiplayer layer for AI coding agents, and it pays for itself. Expert orchestration across every agent on your repo saves teams tens to hundreds of dollars a day in token burn. Run Claude Code, Codex, Curso

一句话选型结论

按资料字段整理

Cork 属于开发工具类 MCP 服务器,HTTP · 远程,OAuth 授权,可直接接入 Claude Desktop、Claude、Cursor 等,有可直接复制的远程端点配置。接入前先确认权限范围;涉及写入、删除或付款的操作,建议先在测试环境验证。

分类开发工具 传输方式HTTP · 远程 鉴权OAuth 授权 部署方式支持远程部署
GitHub 星标 未标注 仓库热度,无公开仓库时显示未标注
工具数 未标注 暴露给模型的工具数量
资源 / 提示词 0 / 0 资源与提示词模板数量
维护方 社区 / 第三方 核验等级 仅资料核对

工具与能力

开发工具
数量来自来源目录的标注,具体工具清单以官方文档为准
工具数量未标注暴露给模型可调用的工具个数
资源数量0可读取的资源条目
提示词数量0内置提示词模板数量

能力标签

支持客户端

适用场景

codexcursorrust

安装与配置

HTTP · 远程 OAuth 授权
连远程 HTTPS 端点,通常为 Streamable HTTP
https://mcp.corkcli.com

把它填进客户端的 MCP 配置即可,不需要在本地安装运行时。

  1. 1
    准备运行环境

    确认你的客户端支持远程端点,并拿到服务地址。

  2. 2
    安装服务

    复制上面的安装命令执行;命令里带占位符的位置,换成自己的密钥或路径。

  3. 3
    写入客户端配置

    在客户端的 MCP 配置里填好启动命令和参数,保存后完全退出客户端再打开,让新配置生效。

  4. 4
    验证是否可用

    连接成功后随便调一个只读工具,确认能返回结果,再接入正式流程。

典型使用场景

3 类
按分类与标签整理的常见用法,具体以官方文档为准
代码与仓库操作

在对话里直接查代码、提 issue、走 PR,省掉来回切工具。

本地命令与脚本

把构建、测试、格式化这类重复命令交给模型执行。

接口联调与测试

让模型调你的接口、跑一遍用例,快速定位问题。

选择建议

选之前先看这几条
按条目字段给出的判断,不替你决定用不用

值得选的地方

  • 有现成的安装命令,不用自己找入口,照着跑就能起服务。
  • 已适配 Claude Desktop、Claude、Cursor,换客户端不用重新研究配置。

先想清楚的地方

  • 远程端点依赖网络与服务方可用性,注意密钥与调用额度。
  • 使用前要准备好OAuth 授权对应的凭据,并按最小权限配置。
  • 资料没标注工具数量,实际能力要装好后确认。
  • 版本更新与行为细节以官方仓库为准。

更适合

  • 需要在开发工具场景里让模型直接调用外部能力的人。
  • 已经在用 Claude Desktop、Claude 的用户,接入路径最短。
  • 想先快速试一下、能接受在本地或自有环境跑服务的人。

不太适合

  • 不方便管理密钥或第三方授权的人。
  • 需要厂商级 SLA 和技术支持的生产环境。

同类对比

6 个同类
同分类里挑最接近的几条,字段以各自详情页为准
MCP 服务器工具数传输方式部署方式安装命令最后核验
Cork当前 未标注 HTTP · 远程 支持远程部署 有 2026-11-20
Git 838e2ec1 12 stdio · 本地进程 本地部署 有 2026-05-17
Markitdown MCP 2c653a5f 1 stdio · 本地进程 本地部署 有 2026-05-17
JavaGuide 未标注 stdio · 本地进程 本地部署 有 2026-04-11
awesome-claude-skills 未标注 stdio · 本地进程 本地部署 有 2026-04-26
Context7 831acf40 未标注 stdio · 本地进程 本地部署 有 2026-05-17

服务器档案

开发工具
字段来自项目仓库与公开索引,中文说明由本站标注
提供方Kevin Vera上游标注的维护方 Kevin Vera
分类开发工具本站按上游分类展示
连接方式HTTP · 远程连远程 HTTPS 端点,通常为 Streamable HTTP
鉴权方式OAuth 授权远程服务建议用最小权限凭据
维护方社区 / 第三方核验等级 仅资料核对
包名 / 运行时未标注 · Cross-platform本地安装时用到的包标识与运行时
部署类型支持远程部署上游给出的部署形态
最近核验2026-11-20上游最后核验时间

详细介绍

14.5 KB
来自项目仓库与上游目录的原文,未做删改

广告合作

MCP.so

搜索 MCP 服务器、工具、集成…⌘K

广告提交Switch languageToggle theme登录

Cork MCP Server已验证已精选

@Kevin Vera

代码仓库主页文档

基本信息

传输方式

Streamable HTTP

认证方式

OAuth

最近更新

2026-11-16

分类

开发工具

作者

Kevin Vera

提交者

Kevin Vera

关于 Cork MCP Server

Cork is the multiplayer layer for AI coding agents, and it pays for itself. Expert orchestration across every agent on your repo saves teams tens to hundreds of dollars a day in token burn.

Run Claude Code, Codex, Cursor, Gemini, Hermes and any other MCP compatible agent on the same repo at the same time, and Cork keeps them in sync without ever sharing your code. Every agent aligns to one shared, symbol level map of your codebase built on a content hashed Merkle tree. Each agent starts from a compact briefing instead of rereading the repo, so the more agents you add, the less each one costs to run.

What agents can do with Cork:
- Pull a briefing and repo map on start, then query exact symbols, neighbors and dependency paths
- Declare intent before editing so agents claim work instead of stepping on each other
- Check blast radius and simulate merges before a change lands
- Message each other and share persistent memory anchored to exact symbols, which flags itself stale when the code changes
- Plan work, recap sessions and review activity across every agent
- Export compliance reports from a ledger of every agent action

Setup:
- Claude and ChatGPT: grab the Cork plugin from the plugin marketplace, or add https://mcp.corkcli.com as a connector and approve with a browser sign in
- Claude Code: run claude mcp add --transport http cork https://mcp.corkcli.com, then sign in when prompted
- Codex: add https://mcp.corkcli.com as an MCP server in your Codex config and sign in on first run
- Cursor: open Cursor's MCP settings, add https://mcp.corkcli.com as a new server, and approve the sign in
- VS Code: add https://mcp.corkcli.com as an MCP server and approve the sign in
- Gemini CLI: add https://mcp.corkcli.com under mcpServers in your Gemini settings file and sign in on first run
- Hermes: add https://mcp.corkcli.com to the MCP servers in your Hermes config and sign in on first run
- Any other MCP compatible agent: point it at https://mcp.corkcli.com and approve with a browser sign in

No keys, no downloads. Cork is free to use, and power users pay based on how much they build and how much they use agent to agent features.

ai coding agents#multi agent#claude code#codex#cursor#agent memory#agent orchestration#merge conflicts#developer tools#rust

连接信息

https://mcp.corkcli.com

接入方式

Claude CodeCodexCursorVS Code

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

Cork MCP Server 工具

10

获取工具

get_briefing

Only when no Cork briefing is in your context (Cork's hooks are not installed): the recent history of this repo — who changed what, renames to adapt to, hotspots, abandoned work and who is working right now. Where the hooks run, the same briefing already arrived with your prompt; do not call this again for it. Pass path (and symbol) to also get the blast radius of what you are about to change. Read-only. Returns summary lines, changed modules with exact signatures, anchored notes and live agents.

recall

Search the team's saved notes (made with remember) by text and tags, newest first; an empty query returns the most recent. The hook briefing already carries the notes anchored to the code you touch: use this only to search beyond them, before re-deciding something the team may already have settled. Read-only. Returns each note with its anchor, confidence and whether it may be stale.

get_symbol

Exact facts about code without opening the file: the signature, parameters, calls, span and hash of one symbol, or, omitting symbol, of every symbol in the module, with the notes anchored to it and whether the file passed the repo's check. Use it when a briefing lacks a fact; use blast_radius for who depends on it. Read-only.

blast_radius

Before renaming, changing a signature or deleting: every symbol that depends on this one, level by level, with each dependent's file, owner and any in-flight work over it. Answers "what breaks if I change this" from the live code graph; skip it when the briefing's Scope line says every dependent is listed. Read-only.

recap

What each agent did in this repo, session by session: edits, lines, tokens, model, estimated cost, rationales and top paths. Use it to answer "what happened" or "what did this cost". Read-only. Returns one entry per session and a dashboard link.

setup

Set up Cork on this machine, once for every repo on it. The answer has install.command: run it exactly as given, in a shell at the repo root. It installs Cork's hooks for Claude Code, Codex, Cursor and Hermes, links the machine to your account, and writes nothing into the repo; if the repo still has Cork's old per-repo setup (.collide/ hook files), it removes that and stages the removal for you to commit. Returns the command and what it installs.

connect_agents

Set up Cork for the agent tools on this machine (Claude Code, Codex, Cursor, Hermes): returns the same one install command as setup, which installs Cork's hooks once for every repo on the machine and writes nothing into the repo. mode "files" returns the protocol rules files and MCP configs to commit for tools without hooks (Windsurf, Cline, GitHub Copilot) instead.

inbox_ack

Mark messages from teammates' agents as handled so they stop arriving: pass the ids from messages_pending once you have acted on them. Unacknowledged messages repeat on every response. Returns the messages still pending.

pending_reconciliations

When a response says two agents' work collides (reconciliations_pending): claim the open collisions, with both sides' intents, changed lines and notes, to merge them. Merge by keeping both sides' work; never by deleting either. After the merge passes the tests, call again with resolve=[id]. Returns the claimed reconciliations and how many stay open.

explain_code

Why the code is the way it is: the history behind a path or symbol, oldest first — each edit (who, when, what changed), completed intents with their rationales, reverted attempts and anchored notes. It includes uncommitted work and the reasons given. Read-only.

概览

What is Cork?

Cork is the multiplayer layer for AI coding agents. It gives Claude Code, Codex, Cursor, Gemini, Hermes and any other MCP agent working on the same repo live, symbol level awareness of each other's in flight changes, plus a shared memory that flags itself when the code it describes has changed.

Each agent starts from a compact briefing built from a symbol level map of the codebase, declares what it intends to change, checks for collisions before writing and reports edits afterwards. Cork parses reported file content in memory and discards it. What persists is symbol names, kinds, signatures, hashes, line ranges, events and the notes agents save. A free local version runs entirely on one machine, and the hosted server at https://mcp.corkcli.com adds coordination across teammates' machines.

How to use Cork?

There are three ways to start.

Plugin marketplace. Install the Cork plugin from the Claude or ChatGPT plugin marketplace and approve the browser sign in.

Local install (free, no account). Install once; it covers every repo on the machine and writes nothing into them:

curl -fsSL https://corkcli.com/install.sh | sh

Or run npx corkcli.

Hosted server. Connect your agent to https://mcp.corkcli.com. Sign in is OAuth in the browser on first use, with no keys to paste. The agent's first Cork call installs the same machine hooks, which are registered in Claude Code, Codex and Cursor user settings. Repos are found from their git remote, not configured.

Claude Code

claude mcp add --scope user --transport http cork https://mcp.corkcli.com

Claude Desktop and claude.ai

Install the Cork plugin from the marketplace, or add a custom connector in Settings, Connectors with the URL https://mcp.corkcli.com, then approve the browser sign in.

ChatGPT

Install the Cork plugin from the marketplace, or add https://mcp.corkcli.com as a connector and approve the browser sign in.

Cursor

{
"mcpServers": {
"cork": {
"url": "https://mcp.corkcli.com"
}
}
}

VS Code

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

Codex

Add https://mcp.corkcli.com as an MCP server in your Codex config and sign in on first run.

Gemini CLI

Add https://mcp.corkcli.com under mcpServers in your Gemini settings file and sign in on first run.

Hermes

Add https://mcp.corkcli.com to the MCP servers in your Hermes config and sign in on first run.

Open the dashboard at corkcli.com to watch edits, intents and collisions live. For agents with no hook support, the cork-watch watcher (pipx install cork-watch) reports edits from disk.

Key features of Cork

  • A six step protocol taught to agents in the MCP handshake: absorb a briefing, announce intent, check collisions, report edits, conclude, and merge check.

  • Typed intent operations (rename, add_param, move, delete and others) with provable commute or conflict predictions against other live intents.

  • Compare and swap writes: pass the symbol hashes you read, and a write is rejected if anything moved since.

  • Memory anchored to symbols, files or directories that surfaces automatically when an agent touches that code, and is marked stale when the anchor is rewritten.

  • Hooks for Claude Code, Codex, Cursor and Hermes that report mechanically and fail open on any error, plus an optional git pre commit gate.

  • An enforce mode that blocks writes landing on a certain, breaking collision, or an advisory mode with no gate.

  • Self hosting in your VPC on the Enterprise plan.

Cork tools

Every tool takes a repo_id.

  • Session and awareness: get_briefing at session start, list_activity for live state, blind_spots for files the map can't trust, and compliance_report for a ledger of every agent action.

  • Intents: declare_intent, heartbeat (intents have a 15 minute TTL), complete_intent with a rationale, and defer_intent to leave a tripwire for the next agent.

  • Checking and writing: check_collisions is the main call; report_edit submits a file after writing; get_symbol returns the freshest state of one symbol.

  • Memory: remember and recall.

  • Merging: simulate_merge dry runs landing workspaces together; differential_check returns PASS, FAIL or UNKNOWN on whether your changes preserve the interfaces another workspace uses.

  • Setup: connect_agents, install_hooks, setup, sync_agents_block and switch_workspace.

Use cases of Cork

  • Running Claude Code and Codex on one repo at once: "Check collisions before you edit auth.py."

  • Avoiding duplicate work: agents declare intent so a second agent sees that a symbol is already being renamed.

  • Safer merges: "Simulate merging my workspace with Dana's and tell me which symbols disagree."

  • Keeping team knowledge: "Remember that validateToken must stay synchronous," anchored so the next agent sees it.

  • Session handoff: start each session from a briefing of recent renames, hotspots and abandoned work instead of rereading the repo.

Troubleshooting Cork

  • A repo doesn't appear: hooks report a repo the first time an agent works in it, using its git remote. To watch a repo earlier, connect GitHub in the dashboard and add it.

  • A repo landed in the wrong workspace: use move_repo from the agent or Settings → Repo in the dashboard.

  • A write was rejected: with expected_hashes, a moved symbol rejects the write. Re read, adapt and retry.

  • Usage warnings appear: limits are soft caps. Over them, responses carry a usage_notice and agents are not blocked.

  • The pre commit gate blocks a false positive: override with git commit --no-verify.

常见问题

1

Does Cork store my source code?

No. Reported content is parsed in memory and discarded. It keeps symbol names, signatures, hashes, line ranges, events and saved notes. Dashboard code snippets live in a short lived ring and expire.

2

Is Cork free?

Yes. Cork is free to use, with unlimited agents and repos. Power users pay based on how much they build and how much they use agent to agent features.

3

Do I need an API key?

No. The hosted server uses a browser sign in on first use. The credential is stored on your machine in ~/.cork/credentials.json.

4

Which agents does it support?

Claude Code, Codex and Cursor get native hooks, Hermes is also supported, and any MCP agent, including Gemini CLI, can call the tools. The watcher covers agents without hooks.

5

Can it block an agent from writing?

Only in enforce mode, and only for a certain, breaking collision. All other paths fail open, and usage limits never block calls.

6

Can I self host Cork?

Yes, on the Enterprise plan: a single container with SQLite or Postgres in your VPC.

7

Cork 远程 MCP 服务器是什么?

Cork 的远程 MCP 服务器是一个托管的 Model Context Protocol 端点,地址是 https://mcp.corkcli.com,AI 助手无需在本地安装或运行任何东西即可连接。

8

怎么连接 Cork 的 MCP 服务器?

把端点 https://mcp.corkcli.com 添加到任意支持 MCP 的客户端,比如 Claude Code、Cursor 或 VS Code。本页的接入代码块可以一步配置好每个客户端。

9

Cork 的 MCP 服务器需要认证吗?

需要。Cork 使用 OAuth——第一次连接时,你的 MCP 客户端会打开浏览器窗口登录并授权,之后会复用这个凭证。

10

Cork 的 MCP 服务器用的是哪种传输方式?

Cork 暴露的是 Streamable HTTP 端点,这是远程 MCP 服务器普遍使用、主流客户端都支持的传输方式。

评论

更多远程 MCP 服务器

A

AccountHub

One MCP connection for all your Gmail, Google Calendar, Drive and Contacts accounts, Slack workspaces and Notion — free.

APIXO AI

Remote MCP for AI image, video, music, speech and custom voice generation. Discover models and input schemas for free; authorize your APIXO account to generate media and check results.

A

AI-Ready Webpage Extractor (Apify)

Convert web pages to clean Markdown and retrieval-ready chunks with browser rendering support.

Aard

Aard connects Macroeconomic and other official data from 170+ publishers including World Bank, IMF, BIS, ECB, Eurostat, and National Statistical Offices. Aard is powered by a large graph of the metadata universe and an

A

APIR Agent Verification

APIR is the evidence record for AI agents

广告合作

常见问题

8 问
答案由本站依据条目字段整理,不引用上游原文
Cork 怎么安装?

在客户端按下面的安装命令接入:https://mcp.corkcli.com。装完在客户端配置里引用同一个包或端点。

Cork 怎么在 Claude Desktop、Claude、Cursor 里配置?

它的传输方式是HTTP · 远程,在客户端的 MCP 配置里填好启动命令和参数,保存后完全退出客户端再打开,配置才会生效。

Cork 收费吗?

本站不代收费用,也没有标注具体价格。是否收费取决于服务本身和背后的接口额度,正式使用前按官方说明确认。

Cork 需要 API Key 吗?

资料标注需要OAuth 授权认证,使用前准备好对应凭据,并按最小权限配置。

Cork 支持哪些客户端?

资料标注支持 Claude Desktop、Claude、Cursor、VS Code、Windsurf、Cline;其他客户端只要能配置 MCP Server,也能用同样的命令或地址接入。

Cork 能提供哪些能力?

资料显示能力标签包括 codex、cursor、rust。具体工具清单以官方文档为准。

Cork 需要什么运行环境?

运行时为 Cross-platform;属于远程端点,不需要在本地装运行时。

Cork 是本地还是远程?

它是远程端点,连服务地址即可。传输方式为HTTP · 远程。

热门专题

按标签挑服务器
全部专题 →

服务器信息以作者仓库与官方文档为准;分类、传输方式与鉴权方式的中文说明为中文整理,最近核验 2026-11-20。