momo-mcp
Allows Momo research to draw on Airbnb lifestyle-service data for market and location research.
Allows Momo research to use Amazon product and pricing data for market, competitor, and product-selection analysis.
Allows Momo research to incorporate App Store reviews and app-market data.
Allows Momo research to use eBay pricing and listing data for market and competitor analysis.
Allows Momo research to incorporate Google Play reviews and app-market data.
Allows Momo research to draw on Reddit discussions and posts as a social-media data source.
Allows Momo research to use Steam reviews and game-market data.
Allows Momo research to use TikTok posts and content as a social-media data source.
Allows Momo research to use YouTube content as a social-media data source.
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
momodi
Momo 决策伙伴 · 你的常驻 AI 调研顾问,现在可以在终端和任何支持 MCP 的 AI Agent 里直接使用。
告诉 Momo 你要做的决策:选品、出海、进入某个市场、评估竞品。它会先澄清范围,再自主完成调研,把来源、假设与下一步建议整理成一份 MBB 级别的研究报告。无需注册可先聊 3 轮。
好决策,从有出处的数据开始:报告里每个关键数据都注明来源,关键数据经交叉验证。
8 大领域实测数据源:TikTok、Reddit、YouTube 等海外社媒原帖;Amazon、eBay、Shein、Lazada 与 30+ DTC 品牌定价;App Store、Google Play、Steam 真实评价;Yahoo Finance、SEC 财报原文与 Polymarket 预测市场;美国海关逐票提单;Booking、Airbnb 等生活服务数据。
先看别人的问题能走多远:Discovery 上有公开的完整调研报告,可以先看研究过程再判断是否适合你。
产品主页 m.moments.top/momo · 公司官网 moments.top · 联系我们 contact@moments.top
这个包把摸摸底的调研能力搬进终端和 AI Agent:一个 npm 包,两个入口,momo 命令行和 momo-mcp MCP 服务,共用一个 HTTP 客户端。
不注册先试
npx -y momodi try "宠物殡葬"不需要账号,匿名 3 轮。用完后会打印注册链接,注册后这段对话会保留。
Related MCP server: DocMind
拿 API key
登录 m.moments.top,打开 /settings/keys,点「生成」。明文只显示一次。
export MOMO_API_KEY=mmd_xxx
npx -y momodi whoami环境变量是唯一配置来源,没有配置文件。可选 MOMO_BASE_URL(默认 https://m.moments.top)。
命令一览
命令 | 参数 | 作用 |
|
| 提交调研并轮询到完成,报告 HTML 写 stdout 或 |
|
| 对已存在的调研接着等 |
|
| 匿名体验,不要 key |
|
| 列出我的报告,最新在前,默认 20 条 |
|
| 取回一份报告;id 可以是 reportId 前缀或 runId |
|
| 浏览公开选题,不要 key |
| 无 | 账号、等级、今日 token、key 前缀 |
--json 时所有命令的 stdout 变为单行 JSON,错误写成 {"error":…} 到 stderr。
典型用法:
momo research "兰州牛肉面出海" --out report.html
momo research "东南亚玩具书" --tier deep --no-wait --json # 拿 runId,稍后 attach
momo reports --json | jq -r '.[].id' | xargs -I{} momo report {} --out {}.html退出码
码 | 含义 | 触发 |
0 | 成功 | |
1 | 未分类错误 | 网络异常、JSON 解析失败、应用 5xx |
2 | 用法错误 | 缺参数、未设 key、 |
3 | 鉴权失败 | 401;403(Momo 权限被停用) |
4 | 调研失败 |
|
5 | 额度不足 | 今日 token 用尽(429);匿名轮数用尽 |
6 | 报告未就绪 |
|
7 | 并发占用 | 已有调研在运行(409) |
130 | 用户中断 | Ctrl-C,调研仍在后台, |
接入 Agent 宿主
包名是 momodi,momo-mcp 是它的第二个 bin,所以 npx 要用 -p 指定包。
# Claude Code
claude mcp add momo -e MOMO_API_KEY=mmd_xxx -- npx -y -p momodi momo-mcp
# Cursor / Codex (.cursor/mcp.json / ~/.codex/config.toml 等价)
{ "mcpServers": { "momo": { "command": "npx", "args": ["-y", "-p", "momodi", "momo-mcp"], "env": { "MOMO_API_KEY": "mmd_xxx" } } } }
# WorkBuddy ~/.workbuddy/mcp.json(macOS 上 command 改为 `which npx` 的绝对路径,改完重启)
{ "mcpServers": { "momo": { "type": "stdio", "command": "/opt/homebrew/bin/npx", "args": ["-y", "-p", "momodi", "momo-mcp"], "env": { "MOMO_API_KEY": "mmd_xxx" } } } }更多排障见 docs/cli.md。WorkBuddy 也可以直接用 momo-research/SKILL.md 这个 Skill 调命令行。
MCP 工具
工具 | 入参 | 返回 |
|
|
|
|
|
|
|
| 报告 HTML 全文 |
|
|
|
Agent 调一次 start_research,之后只在 get_research_status 和等待之间循环;done 后用 reportId 调 get_report。
限制
一个账号同时只能跑一个调研;第二个会得到 409(退出码 7)。批量请串行。
额度按每日 token 配额计,用尽返回 429(退出码 5),次日重置。
--tier只有standard和deep;初判不产出报告,所以不提供。报告格式只有 HTML。
开发
npm install
npm test # vitest,全部对着本地 mock 应用跑,不连真服务
npm run build # tsc → dist/
npm run trace # 从测试标签生成故事追溯矩阵,故事无覆盖时退出 1联系我们
商务与合作:contact@moments.top
公司官网:moments.top
产品主页:m.moments.top/momo
问题反馈:发邮件到 contact@moments.top,或到 GitHub 提 issue
许可
MIT,见 LICENSE。
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
This server cannot be deployed
Maintenance
Related MCP Connectors
MCP tools for agents: web research, content extraction, email, DNS, and blockchain intelligence.
Research briefs and file-based research tasks via MCP. Compare sources, check citations, or analyze authorized data and retrieve reports or files. RigorMed OAuth and available credits are required for task execution.
Your org's AI agents, tasks, runs, search, and brain files as MCP tools and resources.
100+ MCP tools for AI agents: content metadata, trade intelligence, business-expertise analysis.
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceEnables file system operations, web scraping, and AI-powered search through MCP tools for use by LLM agents.1-
- AlicenseNot gradedqualityDmaintenanceAn MCP-native document research agent that provides search, summarization, and citation tools for iterative multi-step research, ending in validated structured JSON reports.MIT
- AlicenseAqualityCmaintenanceEnables AI clients to perform autonomous multi-agent deep research through MCP tools, including deep research, quick search, and retrieval of archived Markdown reports with live web search and source citations.41MIT
- FlicenseAqualityCmaintenanceEnables AI assistants to perform institutional research on companies, financials, industry, news, litigation, and risk, and to generate cited markdown, HTML, or PDF reports through MCP.32-