Model Context Protocol
Name variants MCP · 模型上下文协议 · 模型上下文協議
Get production-ready images, video, audio, and 3D from any AI agent: skills that pick the right model, price before spending, and keep characters and brands consistent through the Scenario MCP, plus expert teams that drive Blender, Maya, ZB
This explanation quotes or summarizes the cited source record. Saving creates a bookmark on this device, with no automatic account sync.
- First observed in our collection
- Oct 10, 2026, 5:00 AM UTC
- Earliest cited publication
- Sep 29, 2026, 1:07 PM UTC
- Last observed
- Nov 5, 2026, 3:30 PM UTC
Evidence in the last complete seven days
- Source mentions
- 25 Previous seven days 32
- Distinct sources
- 14 Previous 21
- Comparable growth
- Not comparable
- Ranking appearances
- 0
Collection observations and undated records are separate from dated publication counts. Observed in 7d:26 · Undated records:6
Collection coverage · Limited historyCurrent / previous observed days 7 / 3
Counts use complete UTC days. First observed means first seen by this collection, not when a term was coined.
Current window:2026-10-29 → 2026-11-05 (end excluded) · History 11 days
| Source | Current days | Previous days | Last success |
|---|---|---|---|
| arXiv | 6 | 3 | Nov 5, 2026, 10:00 AM UTC |
| arXiv | 7 | 2 | Nov 5, 2026, 10:01 AM UTC |
| Ask HN | 7 | 3 | Nov 5, 2026, 3:00 PM UTC |
| Ask HN | 7 | 3 | Nov 5, 2026, 2:30 PM UTC |
| Ask HN | 7 | 3 | Nov 5, 2026, 3:00 PM UTC |
| Ask HN | 7 | 3 | Nov 5, 2026, 3:00 PM UTC |
| GitHub | 7 | 3 | Nov 5, 2026, 12:15 PM UTC |
| GitHub Issues | 7 | 3 | Nov 5, 2026, 10:01 AM UTC |
| GitHub Releases | 7 | 1 | Nov 5, 2026, 12:00 PM UTC |
| GitHub | 7 | 3 | Nov 5, 2026, 2:30 PM UTC |
| GitHub Issues | 7 | 3 | Nov 5, 2026, 10:01 AM UTC |
| GitHub Releases | 7 | 3 | Nov 5, 2026, 3:14 PM UTC |
| GitHub | 7 | 3 | Nov 5, 2026, 2:30 PM UTC |
| GitHub Issues | 7 | 3 | Nov 5, 2026, 10:01 AM UTC |
| GitHub Releases | 7 | 3 | Nov 5, 2026, 3:14 PM UTC |
| GitHub Discovery | 0 | 0 | Nov 5, 2026, 3:13 PM UTC |
| GitHub Trending | 7 | 3 | Nov 5, 2026, 3:14 PM UTC |
| 7 | 2 | Nov 5, 2026, 10:00 AM UTC | |
| Hugging Face | 7 | 2 | Nov 5, 2026, 8:15 AM UTC |
| Hugging Face Spaces | 0 | 0 | Nov 5, 2026, 3:13 PM UTC |
| Hacker News | 7 | 3 | Nov 5, 2026, 12:45 PM UTC |
| Hacker News | 7 | 3 | Nov 5, 2026, 3:30 PM UTC |
| Hacker News | 7 | 3 | Nov 5, 2026, 3:30 PM UTC |
| Hacker News | 0 | 0 | Nov 5, 2026, 3:13 PM UTC |
| Show HN | 0 | 0 | Nov 5, 2026, 3:13 PM UTC |
| MIT News | 7 | 2 | Nov 5, 2026, 10:01 AM UTC |
| OpenAI | 7 | 3 | Nov 5, 2026, 11:30 AM UTC |
| Product Hunt | 7 | 3 | Nov 5, 2026, 12:45 PM UTC |
| 量子位 | 7 | 3 | Nov 4, 2026, 10:15 AM UTC |
| OpenRouter | 0 | 0 | Nov 5, 2026, 3:41 PM UTC |
| Hugging Face Models | 0 | 0 | Nov 5, 2026, 3:41 PM UTC |
| HF Daily Papers | 0 | 0 | Nov 5, 2026, 3:41 PM UTC |
| OpenRouter | 0 | 0 | Nov 5, 2026, 3:41 PM UTC |
| SEC EDGAR | 7 | 3 | Nov 5, 2026, 10:30 AM UTC |
| Sequoia Capital | 7 | 2 | Nov 5, 2026, 10:00 AM UTC |
| Simon Willison | 0 | 0 | Nov 5, 2026, 3:13 PM UTC |
| TechCrunch | 7 | 3 | Nov 5, 2026, 2:30 PM UTC |
| Contracts Finder | 7 | 3 | Nov 5, 2026, 10:00 AM UTC |
| V2EX | 7 | 3 | Nov 5, 2026, 2:30 PM UTC |
- Counts describe monitored public sources, not the whole web or search demand.
- Both weeks need adequate, comparable collection coverage.
Daily history · up to 90 days
Bars show dated source mentions. The latest UTC day may be incomplete. Zero mentions and missing collection coverage are different conditions.
Dotted markers indicate days with no record. They are not counted as zero mentions.
View accessible daily data
| UTC date | Mentions | Sources | Observations | Ranking appearances |
|---|---|---|---|---|
| 2026-09-29 | 1 | 1 | 0 | 0 |
| 2026-09-30 · No record | — | — | — | — |
| 2026-10-01 · No record | — | — | — | — |
| 2026-10-02 · No record | — | — | — | — |
| 2026-10-03 · No record | — | — | — | — |
| 2026-10-04 · No record | — | — | — | — |
| 2026-10-05 · No record | — | — | — | — |
| 2026-10-06 · No record | — | — | — | — |
| 2026-10-07 | 1 | 1 | 0 | 0 |
| 2026-10-08 · No record | — | — | — | — |
| 2026-10-09 | 1 | 1 | 0 | 0 |
| 2026-10-10 | 1 | 1 | 2 | 0 |
| 2026-10-11 | 2 | 2 | 2 | 0 |
| 2026-10-12 | 1 | 1 | 1 | 0 |
| 2026-10-13 | 4 | 3 | 2 | 0 |
| 2026-10-14 | 2 | 2 | 2 | 0 |
| 2026-10-15 | 1 | 1 | 1 | 0 |
| 2026-10-16 · No record | — | — | — | — |
| 2026-10-17 · No record | — | — | — | — |
| 2026-10-18 | 1 | 1 | 0 | 0 |
| 2026-10-19 | 3 | 3 | 2 | 0 |
| 2026-10-20 | 4 | 4 | 4 | 0 |
| 2026-10-21 | 0 | 0 | 2 | 0 |
| 2026-10-22 | 4 | 4 | 4 | 0 |
| 2026-10-23 | 6 | 4 | 6 | 0 |
| 2026-10-24 | 4 | 2 | 4 | 0 |
| 2026-10-25 | 4 | 4 | 6 | 0 |
| 2026-10-26 | 6 | 5 | 6 | 0 |
| 2026-10-27 | 2 | 2 | 3 | 0 |
| 2026-10-28 | 6 | 4 | 5 | 0 |
| 2026-10-29 | 4 | 3 | 4 | 0 |
| 2026-10-30 | 1 | 1 | 2 | 0 |
| 2026-10-31 | 3 | 2 | 2 | 0 |
| 2026-11-01 | 1 | 1 | 2 | 0 |
| 2026-11-02 | 2 | 2 | 2 | 0 |
| 2026-11-03 | 5 | 4 | 4 | 0 |
| 2026-11-04 | 9 | 6 | 10 | 0 |
| 2026-11-05 | 1 | 1 | 8 | 0 |
Source evidence
Sources in the evidence shown:Hacker News 9 · GitHub Trending 2 · V2EX 13 · arXiv 1 · Product Hunt 4 · Simon Willison 1
Hacker News · Published Nov 5, 2026
MCP Explained
View source record →GitHub Trending · Published Not supplied
scenario-labs/skills
Get production-ready images, video, audio, and 3D from any AI agent: skills that pick the right model, price before spending, and keep characters and brands consistent through the Scenario MCP, plus expert teams that drive Blender, Maya, ZB
View source record →V2EX · Published Nov 4, 2026
[分享创造] 为了更舒服的用 PA 搞了点小玩意
让 AI 安全的管你要 token 或发给你 token (适合 Dot / Duet 这种原始的 PA ) http://github.com/imSingee/secure-input 为 PA 增加 Webhook ,以 MCP Events 的方式通知 PA (适合 Dot ) http://github.com/ImSingee/mcp-webhook-events 完全自用 vibe 产品,请让 AI 教你使用/帮你配置(没错,我也不会)
View source record →Hacker News · Published Nov 4, 2026
Show HN: An Open Source Superhuman Email Client Clone
Hi HN, I'm a long-time Superhuman user, and a few weeks ago I cloned it as a native Mac app. It's keyboard-first, with split inboxes, reminders, and a built-in calendar. It's local-first, including the AI features that use local models shipped in macOS 27, with the option to use Claude/Codex subscriptions. It's extremely configurable, even down to the search
View source record →Hacker News · Published Nov 4, 2026
MCP vs. API vs. CLI: How should AI agents connect to your app?
View source record →V2EX · Published Nov 4, 2026
[推广] [开源] Sherlogram:电报福尔摩斯,让 AI 顺着 Telegram 广告追到源头(MCP, 49 个工具)
做了半年 Telegram 灰产调研,把「进群听 → 顺着网址找上游 → 一层层往上爬 → 回 Telegram 核对」这套方法做成了一个 MCP 服务,Claude 等 AI 助手接上就能用。 能干的事:一键溯源出供应链图谱;卖家分 A 到 D 四档并附群里原话;顺着波场 USDT 追钱;挖网站背后的同证书、同服务器、注册时间;骗子库(只收带原话和日期的记录,可导出 Rose 群管封禁格式);每天自动重跑的市场雷达;谁最早发的这条话术;群画像;价格行情;马甲识别;让 AI 以买家口吻去问价。 底层是 49 个 Telegram 工具(读消息、搜全网公开帖、管群、发消息),一份连接全电脑共用。改自 chigwell/telegram-mcp ,Apache-2.0 。 规矩两条:只给证据不下判决;骗子库不收一句
View source record →V2EX · Published Nov 4, 2026
[分享创造] 让 AI 写了个轻量级 MCP 代理 mcp-meta:把 104 个工具折叠成 2 个按需加载,每轮请求 Schema 体积减少 99.4%
https://github.com/monkey-sking/mcp-meta 平时挂多个 MCP Server 时,不管用不用得上,所有工具的完整 JSON Schema 都会在每一轮请求时硬塞进上下文。实测挂 14 个 Server ( 104 个工具) ,模型还没读 Prompt 就先烧掉 14.2 万字符 。 写了 mcp-meta 解决这个痛点。它夹在客户端和任意数量的上游 Server 之间,对模型只暴露 2 个懒加载元工具 : mcp_tools :查看上游 Server 概览,或按紧凑签名(如 search(query: string, [limit: integer]) )按需拉取某个 Server 的工具列表。 mcp_call :指定目标 Server 和工具名发起调用。 接入后每轮请求
View source record →V2EX · Published Nov 4, 2026
[分享创造] [agent 工具]给你的 Agent 接上网页感知与操作能力,送 20 个体验码
大家好,做了大约一年半的 Event Horizon ,简称 EH ,最近把订阅版整理出来了,想请大家实际用一用。 做 Agent 操作网页时,我一直在处理一个问题:任务说得很清楚,但到了页面上,模型能不能找到正确的对象,分清一组内容的归属,再把操作落到对应的控件上? 比如筛选一个列表,分清每张卡片里的标题、价格和按钮;填写表单,找到对应的输入框;遇到弹窗、翻页或页面变化后,再继续往下做。 EH 做的是这层网页感知与操作能力:把当前页面中的可见内容、布局关系和控件整理成结构化的“页面契约”,交给 Agent 。Agent 根据契约决定下一步,再通过工具执行点击、输入、滚动等操作。 使用时不需要先为目标网站编写一套专用操作脚本,而是让 Agent 根据当下的页面继续判断。最终能完成什么任务,也取决于你使用的模型、
View source record →Hacker News · Published Nov 4, 2026
iOS Simulator MCP benchmark: eight tools for AI agents
View source record →Hacker News · Published Nov 4, 2026
AI coding agents are installing abandoned MCP servers with known security issues
View source record →arXiv · Published Nov 4, 2026
MCPGuard: Finding Conformance Gaps in MCP SDKs via Dual-Track Analysis and Dynamic Adjudication
arXiv:2611.03662v1 Announce Type: new Abstract: The Model Context Protocol (MCP) connects large language model agents to external tools and services through official SDKs that implement message validation and session logic. Specification constraints omitted by these SDKs may propagate to downstream applications, highlighting the importance of SDK-level confo
View source record →Product Hunt · Published Nov 3, 2026
backlane
Manage marketplaces in ChatGPT and Claude via MCP
View source record →V2EX · Published Nov 3, 2026
[分享创造] RanklyBee:把 SEO 内容、外链、AI 可见度和获客线索放进一个平台
我们做了一个叫 RanklyBee 的 SEO 工具( https://ranklybee.com ),主要给做网站内容和外链的营销团队、小团队用。 想解决的问题:SEO 的内容、外链、数据分析、AI 搜索可见度,平时要分散在好几个工具里,很难看清全貌。 目前能做的事: 按关键词规划内容日历,写 SEO 文章,并发布到 WordPress 、Webflow 、Shopify 等 CMS 找外链机会,用你自己的邮箱发外联邮件并跟进 追踪 ChatGPT 等 AI 助手有没有提到你的品牌,对比竞品被提及的情况 读取 Search Console 数据,找出值得优化的页面 在社区讨论里发现有购买意向的话题,并生成回复草稿供人工审核 提供 MCP 服务,可以在 Claude 等 AI 助手里直接使用 我是这个产品的开发
View source record →Hacker News · Published Nov 3, 2026
Prax, wallet, transfer and sanctions screening as MCP tools for AI agents
View source record →Product Hunt · Published Nov 3, 2026
Datashake MCP
Plug review and social data straight into your AI assistant
View source record →V2EX · Published Nov 3, 2026
[分享创造] 做了个查询广告案例的 MCP
我一直在广告行业,每次写提案或者策划都需要搜索各种案例,创意枯竭的时候也想看看别人有什么好想法(没错,就是抄作业)。 以前用 Google 搜索,搜出来一堆内容得自己分辨,垂直网站的搜索功能又比较弱。 所以做了一个广告行业的案例库,并提供 MCP 查询功能。 先用爬虫抓了来自各个广告行业案例垂直网站的大概 16 万条内容,烧了十几亿 GPT token 全部分析了一遍,包括匹配重复案例、抽取品牌、创意机制等要素。 最后整理出来 8 万多个案例,19 万个创意机制,覆盖 2 万多品牌。 平时自己也用,感觉查询效果还行,不过实际反馈内容也取决于用户端的 AI 客户端是什么,个人体感 Codex 可能更好一些。 https://admuse.cn/ 注册可以免费查询 3 次,愿意帮忙测试的朋友可以给我注册手机号,我给
View source record →Hacker News · Published Nov 2, 2026
Built open-source auth for AI agents calling MCP servers – looking for builders
View source record →Hacker News · Published Nov 2, 2026
Show HN: Arken – cryptographic identity for AI agents calling MCP servers/APIs
View source record →GitHub Trending · Published Not supplied
beenuar/AiSOC
Open-source AI Security Operations Center: alert fusion, LLM-agent triage, MITRE ATT&CK investigation, and a replayable decision ledger for every agent step. Self-hostable, runs with no API keys, MIT licensed. Ships an MCP server for Cl
View source record →V2EX · Published Nov 1, 2026
[分享创造] 做了个 MCP 工具:给 agent 一个 App Store 链接,它把包拆成可检索的证据
让 agent 分析一个开源项目,它读仓库就行。但一个上架的 App 没有仓库,只有签名后的包:压缩过的 JS 、一个 Mach-O 、一个 plist ,没有入口。"这个竞品怎么做登录的"这类问题,最后变成一下午的解包、grep 、往对话框里贴 JSON 。 我做的就是把那个包变成 agent 能检索、能引用、能推论的证据,通过 MCP 接进你已经在用的客户端。 ( gif 约 2MB ,加载慢的话可以看完整 51 秒版: https://github.com/zhuixin-labs/app-evidence/blob/main/scripts/demo/demo.mp4 ) 终端里是同一套引擎的人类入口: $ apptrace analyze <app-store-link> [resolve
View source record →Product Hunt · Published Oct 31, 2026
MCPCore
Build and host MCP servers with no-code workflows
View source record →V2EX · Published Oct 31, 2026
[分享创造] 分享两个 DSH 小插件:自动加载项目 MCP 和 快捷回复功能
插件一:让 DSH 识别并管理项目级别的 MCP DSH 挺好用的 就是对 MCP 的支持太差了,自己动手 丰衣足食。插件会自动读写项目里 .agents/mcp.json 里的配置 dsh plugin --profile web add dsh-mcp-autoload https://github.com/micln/dsh-mcp-autoload 插件二:快捷回复 现在 AI 在做完一个步骤之后 一般都会给出下一个步骤建议,我就是在不停地发“继续”,但是我现在连继续都懒得打了,就给做成了一个插件 dsh plugin --profile web add dsh-quick-reply https://github.com/micln/dsh-quick-reply 两个插件都是开源的。MCP 已经可以
View source record →V2EX · Published Oct 31, 2026
[推广] [分享创造] send-to-dot:从手机分享图片,接回原来的 ChatGPT 对话
手机上把图片分享给 ChatGPT ,总会新开一个聊天。正在跟 dot 聊到一半,就得找回原对话、再传一次图。 我做了 send-to-dot ,只解决这一件事:从手机分享一张图和一句可选问题,接回你原来在用的 dot 对话,原图作为助手附件发回来,然后再回答问题。 部署在自己的 Mac 上(需持续开机联网),用自己的固定 HTTPS 地址、自己的 ChatGPT 账号通过自建应用( MCP + OAuth )连接。Android 用 Chrome 安装,iPhone 用快捷指令。 隐私上:只接收你主动分享的图片,不扫描相册、不自动同步,也不会冒充你发消息。核心服务不需要 OpenAI 模型 API key ,也不提供作者的主机或额度。 坦白说还在测试中:安装是向导式不是一键,授权默认 30 天需续期,发送是异
View source record →V2EX · Published Oct 30, 2026
[分享创造] 给 AI 做了一个 Dropbox,避免它一换窗口就失忆
平时用 Cursor 、Claude Code 、Codex 做项目的时候,经常遇到一个问题,上下文很容易中断。 比方说我在一个窗口里花了很长时间谈好的项目背景、选好的方案,换个工具或者新开一个会话,基本就没了。要么重新讲一遍,要么自己维护一份越来越长的 MEMORY.md 。 所以我做了一个简单的小工具 Hipo Hipo ,来解决这个问题。 Hipo Hipo 是一个独立于 AI 工具之外的 Context 空间,它会把项目的原始资料、有效决策、约束和方法保存在一起,通过 MCP 连接给 Cursor 、Claude Code 、Codex 、WorkBuddy 等工具使用。不同 Agent 可以读取同一份项目上下文,不需要你每次都从头解释。 Agent 还可以在获得授权后,把新的项目结论写回 Memory
View source record →V2EX · Published Oct 29, 2026
[分享创造] 因为想统一看自己全部券商账户,我做了一个叫 Catfolio 持仓追踪 iOS App
我的美股散落在好几个券商里, 一直找不到一个地方能看清楚: 总共持有什么, 今年到底赚了多少, 跑没跑赢大盘啊。券商 App 只看得到自己那一家, 记账软件要手动录每一笔交易。所以干脆自己写了一个。 支持各种券商 api 连接,不用手动输入(当然也支持自己手动导入 csv )。 买入点卖出点都非常清晰显示在线上,还有金额细节,成本价都算的清清楚楚的。 账户数据都是自动算,合并算分开算,非常直观。 真实收益: 同时算时间加权 (TWR) 和资金加权 (MWR) 收益,可以和 SPY 、QQQ 等基准对比,不会被定投入金带偏。 收益来源: 按标的拆分收益,看清今年的钱是哪几只赚的。 ETF 穿透: 把所有账户里的 ETF 拆开,算出合并后对单只股票的真实敞口。比如 QQQ 、VGT 、VOO 叠在一起,你实际持有多
View source record →Hacker News · Published Oct 29, 2026
Show HN: Fast agentic video editing in the browser (Tella)
Hi HN, today we're launching the biggest update to Tella ever ( https://tella.com/ ). Tella is all-in-one video creation. You can record, edit and publish with one tool. Today we're shipping: 1. Tella Agent: An in-app video editing agent 2. Tella Motion: Rich motion graphics support 3. Auto Capture: Agentic screen recording Our in-app agent uses the same too
View source record →Simon Willison · Published Oct 29, 2026
Quoting Ben Hylak
every product i use should just be a headless mcp, that feeds into an agent, where i never have to think about anything underneath. except for our product, of course. in our product, customers love the ui and love us very much. we're special that way. — Ben Hylak Tags: model-context-protocol , ai
View source record →V2EX · Published Oct 29, 2026
[分享创造] [开源] Shumai,面向 AI 视频创作团队的素材管理与协作平台
v 友们好,想跟大家分享下我最近做的开源项目。先放上直达链接: https://github.com/shumaiOne/shumai 中文 Readme | 官网 | Demo 功能上 Shumai 和 Frame.io / 阅流 / 分秒帧 类似,所以如果你知道他们是干什么的,或者已经在用这些产品了,可以直接去 Demo 站体验一下(注意 Demo 账号是只读的评审员身份)。Shumai 的 UI / UX 以及一些功能都是参考的 Frame.io V4 ,而 阅流 / 分秒帧 则参考的是 Frame.io V3 版本,所以 Shumai 在体验上应该会更现代一些。 相比传统影视拍摄,AI 短剧或漫剧团队往往会在短时间内积累大量的素材,但是大部分 AI 视频生成平台的素材管理功能都比较简陋,更别提协作功能了
View source record →V2EX · Published Oct 28, 2026
[分享创造] 给 Codex 做了块画布,随手试了下,它一笔画完了一只猫
最近在做 Galta ,一块开在网页旁边、可以和 agent 一起用的画布。 一直觉得,这些模型不该只待在聊天框里。把参考、草图和结果放在同一个地方,让它能看到,也能直接动手改,应该会有些不一样的事情发生。 于是随手试了下:把 Pinterest 上一张西装猫的梗图拖进画布,让 GPT-6-Astra 用画笔换个风格画一遍。它画了 620 笔,都是可以继续编辑的矢量笔画。 然后又加了个要求:“再来一次,一笔画完,中途不要抬笔。” 最后是一条包含 70,722 个点的连续笔画。最意外的是,加了这个限制,画出来的猫依然挺像样,效果超过了我的预期。 演示: https://x.com/Koho_Z/status/2115504320712171614 视频由实际笔画数据回放制作,做了剪辑和加速。 画完以后,结果仍然留
View source record →Product Hunt · Published Oct 28, 2026
Suno MCP
The music layerfor your AI stack
View source record →
Review priority 80 / 100
This score orders research candidates. It is not search volume, a forecast or verified commercial demand.
Evidence
30 / 30Distinct dated articles or first observations; repeat crawling adds nothing.
Publisher diversity
25 / 25Conservative owner grouping; syndicated copies count once.
Comparable change
0 / 20Comparable complete weeks only; no percentage from a zero baseline.
Recent evidence
15 / 15Latest publication or first observation, not repeated fetch time.
Name specificity
9.7 / 10Explicit entities and technical concepts outrank ambiguous acronyms.
Limitations and interpretation
- First observation is a collection milestone, not proof that a word or product is new to the world.
- Counts describe monitored public sources, not the whole web or search demand.
- First observed here is not the invention date of a name or idea.
- Candidates come from public titles, summaries and entity URLs, not verified full-text analysis.
- Owner grouping is conservative; undisclosed syndication may remain.
- Undated observations cannot establish publication growth.
- An established concept or common product name retained as a baseline, not a newly invented term.
- Both weeks need adequate, comparable collection coverage.
- Sources behind this term changed or lack a comparable baseline.
- Recent evidence includes collection over three days after publication.