PostIdentity MCP Server

PostIdentity MCP Server

Official MCP server for PostIdentity - Generate AI-powered social media posts, threads, and replies from any MCP-compatible AI assistant with identity management and refinement capabilities.

Category
访问服务器

README

PostIdentity MCP Server

Official MCP server for PostIdentity - Generate AI-powered social media posts from any AI assistant that supports the Model Context Protocol (MCP).

Features

  • 🎭 Manage Identities - List, create, and archive your writing personas
  • ✍️ Generate Posts - Transform thoughts into posts instantly
  • 🧵 Generate Threads - Turn long-form content into cohesive multi-post threads
  • 💬 Generate Replies - Create authentic replies to other posts
  • 🔄 Refine Posts - Regenerate, shorten, lengthen, or adjust style (1 credit per refinement)
  • 📐 Character Limits - Generate posts for Twitter, LinkedIn, or custom lengths
  • 📝 Review Posts - Browse your post history on specific identities
  • 💳 Check Credits - Monitor your credit balance
  • 🎁 Track Referrals - View your referral stats and code
  • 🛍️ Browse Marketplace - Discover identity templates
  • 🔒 Secure - Token-based authentication with RLS protection

Demo

Watch Demo

Click to watch: See PostIdentity MCP in action

Two Deployment Options

🖥️ Local MCP Server (This Package)

For local AI assistants like Windsurf, Claude Desktop, and other MCP-compatible tools.

  • Transport: stdio (stdin/stdout)
  • Installation: npx @postidentity/mcp-server
  • Best for: Desktop AI assistants

☁️ HTTP MCP Server

For cloud-based integrations like Gumloop and other HTTP-based MCP clients.

  • Endpoint: https://izonnqmeqgzmuowhglfx.supabase.co/functions/v1/mcp
  • Transport: HTTP with JSON-RPC 2.0
  • Authentication: X-API-Key header
  • Best for: Cloud platforms, custom integrations

Example usage:

curl -X POST https://izonnqmeqgzmuowhglfx.supabase.co/functions/v1/mcp \
  -H "X-API-Key: pi_xxxxxxxxxxxxx" \
  -H "Content-Type: application/json" \
  -H "Accept: text/event-stream" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'

Both servers provide identical functionality and identical output - just different transport mechanisms.


Prerequisites (Local Server)

  • Node.js 18+ (for local development)
  • A PostIdentity account
  • An MCP-compatible AI assistant (Windsurf, Claude Desktop, etc.)

🚨 v2.0 Migration Notice

If you're upgrading from v1.x: Access tokens still work, but we strongly recommend switching to API keys for permanent access. Access tokens expire hourly and require manual updates.

Quick Start

1. Get Your API Key (Recommended)

  1. Go to postidentity.com/settings
  2. Open the "Developers" section
  3. Click "Create Key"
  4. Give it a name (e.g., "Windsurf MCP")
  5. Copy the key immediately - you won't see it again!

2. Configure Your MCP Client

Recommended (v2.0+):

{
  "mcpServers": {
    "postidentity": {
      "command": "npx",
      "args": [
        "-y",
        "@postidentity/mcp-server@latest",
        "--api-key",
        "your_api_key_here"
      ]
    }
  }
}

Legacy (still supported):

{
  "mcpServers": {
    "postidentity": {
      "command": "npx",
      "args": [
        "-y",
        "@postidentity/mcp-server@latest",
        "--access-token",
        "YOUR_ACCESS_TOKEN_HERE"
      ]
    }
  }
}

⚠️ Note: Access tokens expire every hour and need manual renewal. Use API keys for permanent access.

Available Tools

1. list_identities

Get your writing identities.

Parameters:

  • status (optional): Filter by "active" (default), "archived", or "all"

Examples:

"Show me my PostIdentity identities"
"Show me my archived identities"
"Show me all my identities"

2. generate_post

Transform a thought into a post using an identity's style. Supports refinement options.

Parameters:

  • identity_id (required): Identity UUID or name (e.g., "Tech Blogger")
  • thought_content (required): Your thought/idea
  • character_limit (optional): Max character count (e.g., 280 for Twitter/X)
  • refinement_type (optional): "regenerate", "shorter", "longer", "style_adjust", or "refine"
  • session_id (required for refinements): Random UUID for the refinement session
  • previous_post (required for refinements): The post to refine
  • style_adjustment (for "style_adjust"): "more_casual", "more_formal", "add_humor", "more_serious", "more_direct", "add_emojis"
  • custom_feedback (for "refine"): Your specific feedback

Examples:

"Generate a post as Tech Blogger about AI trends"
"Generate a Twitter post (280 chars) as Tech Blogger about AI trends"

Cost: 1 credit per generation or refinement

3. get_credits

Check your current credit balance.

Example:

"How many PostIdentity credits do I have?"

4. list_posts

Browse your generated posts.

Parameters:

  • profile_id (optional): Filter by identity UUID or name (e.g., "Tech Blogger")
  • limit (optional): Number of posts (default: 10)

Examples:

"Show me my last 5 posts"
"Show me posts from Tech Blogger"

5. get_referral_stats

View your referral code and stats.

Example:

"What's my PostIdentity referral code?"

6. list_marketplace_templates

Browse identity templates in the marketplace.

Parameters:

  • category (optional): Filter by business/creative/personal

Example:

"Show me business identity templates"

7. create_identity

Create a new identity from examples.

Parameters:

  • name (required): Identity name
  • description (required): Writing style description
  • examples (required): Array of 3+ example posts (more = better accuracy)

Example:

"Create a tech blogger identity with these examples: ..."

8. archive_identity

Archive an identity (can be restored later).

Parameters:

  • identity_id (required): Identity UUID or name to archive

Example:

"Archive my Tech Blogger identity"

9. generate_thread

Generate a cohesive thread (multiple connected posts) from one thought. Perfect for breaking down YouTube transcripts, articles, or complex ideas into a series of posts.

Parameters:

  • identity_id (required): Identity UUID or name
  • thought_content (required): Long-form content to transform (YouTube transcript, article, etc.)
  • post_count (optional): Number of posts (3-10) or "auto" (default: "auto")
  • platform (optional): "twitter" (280 chars) or "linkedin" (500 chars) - default: "twitter"
  • character_limit (optional): Override character limit per post

Examples:

"Generate a Twitter thread as Tech Blogger about this YouTube transcript: [paste transcript]"
"Create a 5-post LinkedIn thread as Tech Blogger about AI trends"

Cost: 1 credit per thread (regardless of number of posts)

10. generate_reply

Generate a reply to an existing post in your identity's voice. Perfect for engaging with other users' content authentically.

Parameters:

  • identity_id (required): Identity UUID or name
  • original_post (required): The post you're replying to (full text)
  • reply_direction (required): Your intended points/reaction
  • character_limit (optional): Max characters (default: 280)

Examples:

"Generate a reply as Tech Blogger to this post: [paste post]. Agree with them but add nuance about AI safety."
"Reply as Tech Blogger: [original post]. Point out a flaw in their argument about productivity tools."

Cost: 1 credit per reply

Usage Examples

List Your Identities

You: "List my PostIdentity identities"
AI: "Found 2 identities:
     1. Tech Thought Leader
        ID: abc123...
     2. Casual Developer
        ID: def456..."

Generate a Post

You: "Generate a post as Tech Thought Leader about AI safety"
AI: "✅ Post generated successfully!
     
     AI safety isn't about preventing sci-fi scenarios.
     It's about ensuring predictable behavior TODAY.
     
     💳 Remaining credits: 45"

Refine a Post

You: "Make that post shorter" (with refinement parameters)
AI: "✅ Post generated successfully!
     
     AI safety is about predictable behavior TODAY,
     not sci-fi scenarios.
     
     💳 Remaining credits: 44"  (1 credit charged)

Check Credits

You: "How many credits do I have?"
AI: "💳 Current Balance: 45 credits
     ✅ You can generate 45 more posts"

View Referral Stats

You: "What's my referral code?"
AI: "🎁 Referral Stats
     📋 Your Code: ABC123XY
     👥 Total Referrals: 3
     💰 Credits Earned: 15"

Generate a Thread from YouTube Transcript

You: "Generate a Twitter thread as Tech Blogger about this YouTube transcript:
      [paste long transcript here]"
AI: "✅ Thread generated successfully!
     
     📝 5 posts in thread
     ──────────────────────────────────────────────────
     
     **Post 1/5** (276 chars)
     [Hook: Attention-grabbing opening]
     ──────────────────────────────
     
     **Post 2/5** (280 chars)
     [Development: Key point 1]
     ...
     
     **Post 5/5** (265 chars)
     [Conclusion: Strong takeaway]
     
     💳 Remaining credits: 44
     🆔 Thread ID: abc123..."

Generate a Reply

You: "Reply to this post as Tech Blogger: 'AI will replace all programmers by 2030'
      Disagree politely but firmly, emphasizing augmentation over replacement."
AI: "✅ Reply generated successfully!
     
     📩 **Original post:**
     AI will replace all programmers by 2030...
     
     💬 **Your reply:**
     ──────────────────────────────────────────────────
     Respectfully disagree. The last 20 years of 
     automation didn't replace programmers—it made 
     them 10x more productive. AI is a tool, not a 
     replacement. The best devs will be those who 
     master the tool.
     ──────────────────────────────────────────────────
     
     💳 Remaining credits: 43"

Local Development

# Clone the repository
git clone https://github.com/PostIdentity/mcp-server.git
cd mcp-server

# Install dependencies
npm install

# Build
npm run build

# Run locally
node dist/index.js --access-token YOUR_TOKEN

Security

✅ Token-based authentication - No passwords stored
✅ Row Level Security (RLS) - Database-level access control
✅ User isolation - Can only access your own data
✅ Credit system - Natural rate limiting
✅ Audit trail - All transactions logged

Troubleshooting

Tools Not Appearing

  1. Verify JSON syntax in config file
  2. Check token is correct and not expired
  3. Restart your AI assistant completely
  4. Check logs for errors

Authentication Errors

  1. Get fresh token from postidentity.com/settings
  2. Update your MCP config with new token
  3. Restart AI assistant

"Insufficient Credits" Error

Buy credits at postidentity.com/credits

How It Works

  1. You provide your access token in the MCP config
  2. Server authenticates using token on startup
  3. Token is used for all API requests to PostIdentity
  4. All operations respect your credits and RLS policies

Credits & Pricing

  • Free Trial: 5 credits on signup
  • Generation: 1 credit per post
  • Refinements: 1 credit per refinement
  • Referrals: 5 credits per referral (for both users)
  • Pricing: Starting at $5 for 15 credits

Buy credits →

Refinement Features

Refine your generated posts with these options:

  • regenerate: Generate a completely new version
  • shorter: Reduce length by ~40% while keeping the message
  • longer: Add more detail and context
  • style_adjust: Apply a single predefined tone change (more casual, formal, humor, etc.)
  • refine: Apply any custom feedback, mixed instructions, or specific edits

How It Works: Your AI assistant automatically chooses the right refinement type based on your natural language request:

  • "Make it shorter" → shorter
  • "Add more humor" → style_adjust: add_humor
  • "Make it more direct and remove em dashes" → refine (custom feedback for mixed instructions)
  • "Try again" → regenerate

Note: In the MCP server, all refinements cost 1 credit. The web app at postidentity.com offers the first 3 refinements free per session.

Links

Contributing

Contributions are welcome! Please open an issue or submit a PR.

License

MIT © PostIdentity


Made with ❤️ by the PostIdentity team

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选