M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
P

Prompts.chat

Community
The world's largest open-source prompt library for AI
How to install ↓ Compare GitHub source repository ↗
Category
Other
Stars
★ 173.4k Very popular
Transport
stdio (local process) · Streamable HTTP
Runtime
Hosted (no local runtime) · Node.js
Credentials
No credential needed
License
Other / unspecified
Last commit
Tools
1
31FMRS · D

prompts.chat is a mature open-source prompt library with a massive community and wide recognition. As an MCP server, it offers both remote and local deployment, but specific tools are not publicly documented. It is suitable for developers and teams needing high-quality prompts, but note the potential risks of prompt content.

Strongest · Setup experience 8/20 Weakest · Reliability 3/20

Reliability
3/20
Security and permissions
5/20
Maintenance
8/20
Documentation
7/20
Setup experience
8/20
Why each score
Reliability 3/20
Evidence: README advertises an MCP server (remote URL https://prompts.chat/api/mcp and local `npx prompts.chat mcp`) and provides client config snippets. No server manifest, source, tests, or CI workflow was supplied, so the init/tool-list handshake and tool behavior cannot be verified. Deducted heavily for missing execution evidence and unverifiable claims; cannot award above the low end per the static calibration.
Security and permissions 5/20
Evidence: no credentials or secrets appear in the README; the service is a read-only prompt library and no dangerous operations (delete/pay/deploy) are advertised. However, without source code, data-flow, network egress, and permission boundaries cannot be audited; no explicit confirmation flows or least-privilege analysis is visible. Deducted for missing scoping/telemetry evidence; no red-line violation found.
Maintenance 8/20
Evidence: repository is not archived, has open issue tracker, recent README additions (MCP server, book, kids section), contributor graph, sponsors, and a dual MIT/CC0 license statement. However, no release tags, commit history, dependency update cadence, or security response channel were provided; GitHub license field shows NOASSERTION. Deducted for governance/versioning gaps.
Documentation 7/20
Evidence: README provides two MCP setup examples (remote URL and local npx command), links to MCP API docs and self-hosting/docker guides. Missing: explicit tool list, parameter schemas, rate limits, cost, auth requirements for MCP, and troubleshooting steps within the supplied material; external doc content was not provided. Deducted for hidden assumptions and unverifiable linked docs.
Setup experience 8/20
Evidence: setup path is clearly described: remote config is a one-line URL addition; local uses `npx -y prompts.chat mcp`. No runtime prerequisites (Node version), platform matrix, or verification steps are given, and no CI/tests in the supplied files confirm the command works. Deducted for unverified command behavior and missing compatibility details; per static calibration, cannot exceed 15.

Static review · not runListed 2026-08-07

Read the FMRS scoring method →

Fit and risk

What it can accessUses the network

Best for

  • AI users seeking high-quality prompts quickly
  • Teams wanting to build an internal prompt library
  • Learners interested in prompt engineering
  • Educators (for kids' prompt learning)
  • Enthusiasts of open-source and community-driven projects

Not for

  • Enterprises needing a dedicated private prompt management tool (unless self-hosted)
  • Users looking for specific proprietary prompt sets
  • Users needing fully offline functionality (remote mode requires network)

Required permissions

  • May require network access for remote MCP mode
  • Local mode requires Node.js and npm installed (via npx)
  • Self-hosting requires a database (recommended PostgreSQL) and optional authentication configuration

Risks and side effects

  • Prompts may contain biased or inaccurate content (community-submitted)
  • User prompts may be unfiltered; discretion advised
  • Self-hosting requires managing database and authentication, which may involve security configuration risks

Setup

Before you start

Runtime：Hosted (no local runtime) · Node.js

To use via MCP, there are two options:

  1. Remote (recommended): Add the remote URL in your MCP client config:
{
  "mcpServers": {
    "prompts.chat": {
      "url": "https://prompts.chat/api/mcp"
    }
  }
}
  1. Local: Run the MCP server via npx:
{
  "mcpServers": {
    "prompts.chat": {
      "command": "npx",
      "args": ["-y", "prompts.chat", "mcp"]
    }
  }
}
MCP config · JSON 
{
  "mcpServers": {
    "prompts.chat": {
      "url": "https://prompts.chat/api/mcp"
    }
  }
}

Works in clients that read an "mcpServers" key. For clients such as VS Code that use a different key, the configurator below converts it.

.vscode/mcp.json
{
  "servers": {
    "prompts.chat": {
      "url": "https://prompts.chat/api/mcp"
    }
  }
}

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal
claude mcp add --transport http prompts.chat https://prompts.chat/api/mcp

Run it in a terminal; replace any <…> placeholders with your own values first.

Generate and validate client config

Check that it works

Configure https://prompts.chat/api/mcp in your MCP client and reconnect; the prompts.chat tools should appear in the client's tool list, and calling any of them and receiving prompt content confirms the connection works.

Troubleshooting

  1. Remote connection fails: Check network and MCP endpoint URL
  2. Local npx command fails: Ensure Node.js and npm are globally installed with compatible versions
  3. Self-hosting issues: Check PostgreSQL database connection and setup script success
  4. MCP tool list empty: Confirm the server started and connected properly

Things to try

Once connected, you can ask your AI assistant things like:

  • Search the community prompt library for writing-related role prompts
  • Recommend a few prompts from prompts.chat useful for programmers
  • Find a prompt in prompts.chat for converting text into SQL queries

Tools 1

unknown
Prompt library tool (exact tools not documented in source)

Use cases

Discover and browse community-contributed AI prompts
Quickly use curated prompts with any AI chat model
Self-host a private prompt library for organizational sharing
Learn prompt engineering via the interactive book
Teach kids AI communication through gamified adventure

Overview

prompts.chat (formerly known as Awesome ChatGPT Prompts) is a community-driven, free and open-source prompt library for discovering, sharing, and collecting prompts. It works with ChatGPT, Claude, Gemini, Llama, Mistral, and more, and can be accessed via the website, GitHub, or as an MCP server. The project has 143k+ GitHub stars, was featured by Forbes, referenced by Harvard and Columbia, and is the most-liked dataset on Hugging Face.

Similar servers

Telegram MCP Server 70 · B

Telegram MCP server powered by Telethon to let MCP clients read chats, manage groups, and send/modify messages, media, contacts, and settings.

★ 1.8k · Tools 76 Compare with this →
Compare with other MCP servers

Source revision de07abaca567 Data synced 2026-10-19 Read the FMRS scoring method