Schedulala

com.schedulala由 schedulala.com 发布官方发布者v0.10.1更新于 Nov 2, 2026

Schedule and publish social posts to 12 platforms with first comments, analytics, and listening.

已验证Streamable HTTP可网页运行沟通与协作数据与分析

概览

AI 生成的概览

让 AI 助手在 12 个平台上排期、发布和分析社交媒体帖子,并支持首条评论、评论处理和关键词监听。

可以做什么
  • 在 12 个平台(含 X、Instagram、TikTok、LinkedIn、YouTube、Facebook、Threads、Bluesky、Pinterest、Mastodon、Telegram、Google 商家资料)排期并发布帖子
  • 通过一个 API 字段在 8 个平台上添加首条评论
  • 发布前按各平台的字符限制和媒体规则校验草稿
  • 查看已发布帖子的数据分析和最佳发布时间
  • 在同一对话中处理评论并进行关键词监听
  • 在托管连接器中使用帖子预览、媒体上传、日历和收件箱等交互组件
适用场景
适合希望在一个对话中端到端运营社交媒体的场景:跨多平台起草、校验、排期和发布,然后查看表现并回复评论。也适合需要由服务器代为处理各平台发布规则的智能体。
运行要求
可使用远程端点并通过 OAuth 登录(无需处理密钥),或使用本地 npm 包 @schedulala/mcp-server(stdio 传输),后者需要 Node.js 和 SCHEDULALA_API_KEY 环境变量。需要 Schedulala 账号和 API 密钥;免费密钥含 2 个已连接账号和 4 条帖子,sk_test_ 密钥可模拟而不实际发布。托管连接器需要付费 Claude 套餐。
安装前请注意
该服务器会向真实社交账号发布内容,帖子和评论一旦发出即公开且难以撤回,建议先用 sk_test_ 密钥模拟。它需要具有发布权限的有效 API 密钥(SCHEDULALA_API_KEY)。发布会计入套餐额度,可能涉及付费方案。已连接账号和帖子内容会发送给 Schedulala。

安装

在 SourceWeft 中

  1. 打开 控制台中的 Schedulala,将其添加到工作区。
  2. 连接 所需的 凭据。它们会被私密存储,绝不会在公开页面上展示。
  3. 为需要使用其工具的对话启用该服务。

Web executable,通过 Streamable HTTP。 远程服务在工作区中配置后即可从网页运行时运行。

交给智能体安装

把这段提示词粘贴到 Claude Code、Codex、Cursor 等智能体,或 SourceWeft 对话中。智能体会读取该服务器的安装指南,先向你展示发布者、端点和登录方式,经你同意后再把它添加到你的客户端。密钥不会写进任何命令。

阅读 https://sourceweft.com/mcp/com.schedulala%2Fschedulala/install.md,并按其中说明把这个 MCP 服务器接入我的客户端。修改任何配置之前,先告诉我它的发布者、端点和登录方式,等我确认。不要把 API 密钥写进命令或我会提交的文件;问我密钥应从哪里获取。

查看智能体所依据的安装指南

其他 MCP 客户端

把它添加到你客户端的 mcpServers 配置中。 客户端会在首次连接时完成 OAuth 登录。

{
  "mcpServers": {
    "schedulala": {
      "type": "http",
      "url": "https://schedulala.com/api/mcp"
    }
  }
}

README

[Schedulala logo]

Schedulala

Your agent posts. Replies. Measures. Listens.

12 Platforms. Any Agent. One Prompt.

[npm CLI] [npm MCP server] [License MIT] [MCP Registry] [X Follow]


Schedulala teaches any AI agent to run your social media end to end: schedule and publish to 12 platforms — Twitter/X, Instagram, TikTok, LinkedIn, YouTube, Facebook, Threads, Bluesky, Pinterest, Mastodon, Telegram, and Google Business Profile — with first comments on 8 of them through one API field, every post validated against each platform's character limits and media rules before anything publishes, plus analytics, comment triage, and keyword listening from the same conversation. The skill was tested against genuinely cold agents until one installed, connected, and posted with zero human help.

Quick start

Works with OpenClaw, Hermes, Claude Code, Cursor, Codex, claude.ai, ChatGPT, and anything with a shell. Per-runtime guides: schedulala.com/agents.

CLI · MCP (remote or stdio) · sandbox keys that never publish

Get your API key from schedulala.com, then send this message to your agent:

Read https://schedulala.com/SKILL.md and follow it to set up Schedulala. My API key is sk_live_·········

That's the whole install. The hosted skill walks your agent through the CLI (npm i -g schedulala, zero tool-schema tokens), or MCP if you prefer typed tools — and teaches the safe workflow: draft → validate/preview → confirm → publish → verify, plus the per-platform rules that reject posts (TikTok privacy + postMode, Instagram media, YouTube titles, character limits, …).

No key yet? A free one takes two minutes (npx schedulala init --email [email protected]) and includes 2 connected accounts and 4 posts to try the loop. sk_test_ keys simulate everything without publishing.

The hosted connector (claude.ai / Claude Desktop / ChatGPT)

No install at all — interactive widgets included (post previews, in-chat media upload, analytics cards, calendar, inbox). In Claude: Settings → Connectors → Add custom connector and paste:

https://schedulala.com/api/mcp

OAuth sign-in, no API key handling. Requires a paid Claude plan for custom connectors; the same URL works in ChatGPT. Setup guide: https://schedulala.com/developers/docs/claude-connector

[Connecting Schedulala to Claude: Settings → Connectors → Add custom connector → paste https://schedulala.com/api/mcp — set up in two minutes]    [Posting from one chat: drop food photos into the upload widget, the agent drafts 7 per-platform posts with previews, then schedules them across the week]

Set up in two minutes — then post everywhere from one chat.
Full 45s demo (sound on) · Setup · Posting · Comments · Bulk · Best times · Analytics

Install the skill locally (optional)

The one-liner above needs no install. If your runtime prefers a local skill:

bash
# skills CLInpx skills add schedulala/schedulala-agent
# Claude Code plugin marketplace/plugin marketplace add schedulala/schedulala-agent
# OpenClaw (ClawHub)openclaw skills install schedulala
# plain gitgit clone https://github.com/schedulala/schedulala-agent ~/.claude/skills/schedulala-agent

Pair it with either:

  • the local MCP server — npx -y @schedulala/mcp-server with SCHEDULALA_API_KEY (Claude Desktop / Claude Code / Cursor config in the package README), or
  • the CLI — npx schedulala init --email [email protected], then npx schedulala post "Hello" --platforms twitter --now. Every command supports --json; exit codes 0–7.

What's in this repo

skills/schedulala/SKILL.md      the skill: install, workflow, rules, decision treesskills/schedulala/references/   per-platform schemas, media protocol, CLI, widgetsexamples/                       ready-to-adapt JSON payloads (tiktok, threads, bulk, …).claude-plugin/                 Claude Code plugin + marketplace manifest

Links

License

MIT — see LICENSE. The skill documents the Schedulala service, which is a separate commercial product.

来源:README.md,提交 9f7c3f0

工具

0
该服务器需要登录,因此 SourceWeft 无法在没有账号的情况下读取其工具,也未收录。已连接账号所能看到的内容可能因账号及其权限而异。

版本历史

1
  1. v0.10.1最新Sep 16, 2026

评分

已满足适用检查项的 85%

该评分衡量此收录条目的完整程度,依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价,也不表示服务器安全或当前可用:可连通性是带日期的检查结果,评分也不包含安装量或用户评价。

  • 近 30 天内可连通必需25 分
  • 已声明认证方式必需10 分
  • 已列出工具15 分

    请让服务器在无凭据时响应 tools/list,或发布可在沙箱中无凭据启动的软件包。

  • 描述必需15 分
  • 图标10 分
  • 文档15 分
  • 许可证10 分