Publer MCP Server
Wraps the Publer social media scheduling API to let you manage posts from Claude, OpenAI Codex CLI, and Gemini CLI.
README
Publer MCP Server
An MCP (Model Context Protocol) server that wraps the Publer social media scheduling API, letting you manage posts from Claude Code, Claude Desktop, OpenAI Codex CLI, and Gemini CLI.
Quick start
npm install
npm run setup # interactive wizard
The wizard will:
- Ask for your Publer API token and validate it live
- Build the server (
tsc) - Let you pick which agents to configure
- Write the correct config file for each agent
Getting your Publer API token
- Log in at https://publer.com/paweltkaczyk
- Profile → Settings → API
- Click Generate Token and copy the value
Manual installation
If you prefer to configure agents by hand, build first:
npm run build
Then add the following snippet to each agent's config file, replacing the values:
{
"mcpServers": {
"publer": {
"command": "node",
"args": ["/absolute/path/to/publer-mcp/dist/index.js"],
"env": {
"PUBLER_API_TOKEN": "YOUR_TOKEN_HERE"
}
}
}
}
Config file locations
| Agent | Config file |
|---|---|
| Claude Code | ~/.claude/settings.json |
| Claude Desktop (macOS) | ~/Library/Application Support/Claude/claude_desktop_config.json |
| Claude Desktop (Windows) | %APPDATA%\Claude\claude_desktop_config.json |
| Claude Desktop (Linux) | ~/.config/Claude/claude_desktop_config.json |
| OpenAI Codex CLI | ~/.codex/config.json |
| Gemini CLI | ~/.gemini/settings.json |
Available tools
| Tool | Description |
|---|---|
get_current_user |
Get the profile of the currently authenticated Publer user |
list_workspaces |
List all Publer workspaces the API token has access to |
list_accounts |
List all connected social media accounts in the current workspace |
list_posts |
List posts with optional filters (state, date range, etc.) |
get_post |
Retrieve a specific post by ID |
update_post |
Update an existing post's text, schedule, or media |
schedule_post |
Create a scheduled, draft, or recurring post across accounts |
publish_post_now |
Publish a post immediately across one or more accounts |
delete_post |
Delete a post by ID |
get_post_insights |
Get performance metrics for published posts |
get_best_times |
Get optimal posting times based on audience activity |
list_media |
List media assets in the workspace library |
upload_media_file |
Upload a local file to the media library |
upload_media_from_url |
Upload media from a public URL |
get_job_status |
Poll the status of an asynchronous job (e.g. post creation) |
get_social_manager_instructions |
Get high-level skill instructions and workflows |
split_content_into_thread |
Intelligently split long text for threads |
validate_post |
Check a post against platform constraints |
manage_account_presets |
List, create, or delete groups of accounts |
schedule_posts_bulk |
Schedule multiple posts in a single request |
cleanup_media |
Bulk delete media assets from the library |
Advanced Features
- Account Presets: You can group multiple social media account IDs into a single label (e.g.,
@product). Once created viamanage_account_presets, you can use@labelin anyaccount_idsfield. - Follow-up Comments: Tools like
schedule_postandpublish_post_nowsupport afollow_up_textparameter. On platforms like X, Threads, and Mastodon, this automatically creates a thread. On Facebook and LinkedIn, it adds a comment. - Social Manager Skill: This repository includes specialized agent instructions for handling platform-specific character limits, intelligent thread splitting, and optimal posting time selection.
- Gemini CLI: Install as a native skill via
npm run setup. - Claude Code: Setup automatically installs these as a native skill to
.claude/skills/publer-social-manager/. - All Agents: Can access instructions by calling the
get_social_manager_instructionstool.
- Gemini CLI: Install as a native skill via
Development
npm run dev # run with tsx (no build step)
npm run build # compile to dist/
npm run setup # re-run the wizard (e.g. to rotate token or add agents)
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。