Prompts.Chat MCP Server
- 32 repo stars
- Updated March 28, 2026
- f/prompts.chat-mcp
io.github.f/prompts.chat-mcp is a MCP server that searches and retrieves AI prompts from prompts.chat inside your agent.
About
prompts.chat MCP lets developers query a social catalog of AI prompts from inside their coding agent instead of context-switching to the web app. Install @fkadev/prompts.chat-mcp, connect via stdio, and retrieve prompts that match filters you pass through PROMPTS_QUERY for users, categories, or tags. Add PROMPTS_API_KEY when you want write access to save prompts back to the community. It suits developers shipping features with Claude Code, Cursor, or Codex who reuse battle-tested prompt patterns for support bots, content tools, or internal agents. It does not replace your own SKILL.md workflows or local prompt files—it is a live directory integration. Version 1.0.9 is published with documented optional and filter environment variables.
- Search and retrieve prompts from the prompts.chat social platform
- Optional PROMPTS_API_KEY enables authenticated save_prompt tooling
- PROMPTS_QUERY filter string for users, categories, and tags
- npm package @fkadev/prompts.chat-mcp v1.0.9 over stdio
Prompts.Chat MCP Server by the numbers
- Data as of Jul 7, 2026 (Skillselion catalog sync)
claude mcp add --env PROMPTS_API_KEY=YOUR_PROMPTS_API_KEY --env PROMPTS_QUERY=YOUR_PROMPTS_QUERY prompts.chat-mcp -- npx -y @fkadev/prompts.chat-mcpAdd your badge
Show developers this MCP server is listed on Skillselion. Paste this into your README.
| repo stars | ★ 32 |
|---|---|
| Package | @fkadev/prompts.chat-mcp |
| Transport | STDIO |
| Auth | Required |
| Last updated | March 28, 2026 |
| Repository | f/prompts.chat-mcp ↗ |
What it does
Search and pull community AI prompts from prompts.chat while coding, and optionally save prompts when an API key is set.
Who is it for?
Best when you're iterating on LLM features and want a searchable social prompt library on demand in the IDE.
Skip if: Skip if you forbid external prompt sources, need offline-only assets, or do not use MCP in your agent stack.
What you get
Your agent can fetch filtered prompts from prompts.chat and optionally save new ones when an API key is configured.
- In-session prompt search results from prompts.chat
- Optional saved prompts when API key is set
By the numbers
- Server version 1.0.9
- Transport: stdio
- registryType: npm @fkadev/prompts.chat-mcp
README.md
prompts.chat MCP Server
Access thousands of AI prompts directly in your AI coding assistant
Features
- MCP Prompts - Browse and use prompts directly via MCP prompts capability
- Search Prompts - Search for AI prompts by keyword, category, or tag
- Get Prompt - Retrieve prompt details by ID with variable substitution
- Variable Support - Prompts with
${variable}syntax are automatically handled
🛠️ Installation
Requirements
- Node.js >= v18.0.0
- An MCP-compatible client (Cursor, Windsurf, VS Code, Claude Code, etc.)
Install in Cursor
Add this to your Cursor MCP config file (~/.cursor/mcp.json):
Remote Server (Recommended)
{
"mcpServers": {
"prompts-chat": {
"url": "https://prompts.chat/api/mcp"
}
}
}
Local Server
{
"mcpServers": {
"prompts-chat": {
"command": "npx",
"args": ["-y", "@fkadev/prompts.chat-mcp"]
}
}
}
Install in Windsurf
Add this to your Windsurf MCP config file:
Remote Server (Recommended)
{
"mcpServers": {
"prompts-chat": {
"serverUrl": "https://prompts.chat/api/mcp"
}
}
}
Local Server
{
"mcpServers": {
"prompts-chat": {
"command": "npx",
"args": ["-y", "@fkadev/prompts.chat-mcp"]
}
}
}
Install in VS Code
Add this to your VS Code MCP settings:
Remote Server (Recommended)
"mcp": {
"servers": {
"prompts-chat": {
"type": "http",
"url": "https://prompts.chat/api/mcp"
}
}
}
Local Server
"mcp": {
"servers": {
"prompts-chat": {
"type": "stdio",
"command": "npx",
"args": ["-y", "@fkadev/prompts.chat-mcp"]
}
}
}
Install in Claude Code
Remote Server (Recommended)
claude mcp add --transport http prompts-chat https://prompts.chat/api/mcp
Local Server
claude mcp add prompts-chat -- npx -y @fkadev/prompts.chat-mcp
Install in Zed
Add this to your Zed settings.json:
{
"context_servers": {
"prompts-chat": {
"command": {
"path": "npx",
"args": ["-y", "@fkadev/prompts.chat-mcp"]
}
}
}
}
⚙️ Configuration
The local server supports the following environment variables:
| Variable | Description |
|---|---|
PROMPTS_API_KEY |
Optional API key for authenticated requests |
PROMPTS_QUERY |
Optional query string to filter prompts (e.g., users=a,b&categories=c,d&tags=e,f) |
Example with Environment Variables
{
"mcpServers": {
"prompts-chat": {
"command": "npx",
"args": ["-y", "@fkadev/prompts.chat-mcp"],
"env": {
"PROMPTS_API_KEY": "your-api-key",
"PROMPTS_QUERY": "users=username&categories=coding&tags=productivity"
}
}
}
}
🔨 Available Tools
| Tool | Description |
|---|---|
search_prompts |
Search for AI prompts by keyword. Supports filtering by type, category, and tag. |
get_prompt |
Get a prompt by ID. Supports variable elicitation for prompts with template variables. |
save_prompt |
Save a new prompt to your account. Requires PROMPTS_API_KEY. |
📚 MCP Prompts
This server exposes all public prompts from prompts.chat as MCP prompts. Use prompts/list to browse available prompts and prompts/get to retrieve them with variable substitution.
📖 Example Usage
Ask your AI assistant:
Search for prompts about code review
Get the prompt for "act as a linux terminal"
🔗 Links
- prompts.chat - Browse all prompts
- API Documentation - API reference
- GitHub - Source repository
📄 License
ISC
Recommended MCP Servers
0Latency MemoryPersistent memory layer for AI agents.
0nMCP — Universal AI API OrchestratorUniversal AI API Orchestrator — 1,554 tools, 96 services.
0xHumans Protocol MCPMCP for AI agents: financing, skills, lending on Base1k Patient Mcp1k patient MCP server
1trippulse1trip PULSE: 21-tool AI travel planner.
3D AI Agent Avatar3D AI Agent Avatar — render any GLB, give it a Solana wallet, a voice, and pump.fun powers.89How it compares
Live prompts.chat directory MCP, not a local SKILL.md pack or code review tool.
FAQ
Who is io.github.f/prompts.chat-mcp for?
Developers using MCP-enabled agents who want prompts.chat discovery and retrieval without leaving the repo.
When should I use io.github.f/prompts.chat-mcp?
Use it during build when you need vetted community prompts for features, evals, or agent instructions and may want to publish saves with an API key.
How do I add io.github.f/prompts.chat-mcp to my agent?
Add stdio MCP config pointing at @fkadev/prompts.chat-mcp, set PROMPTS_QUERY as needed, and optionally PROMPTS_API_KEY for save_prompt.