# rtcf [Health: Active]

**Category:** 🎥 Multimedia Process  
**Repository:** https://github.com/hengkp/rtcf-mcp  
**GitHub Stars:** 0  
**Views:** 0  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/rtcf

## Description
Turns rough requests into sharp Role/Task/Context/Format prompts. Thai and English.

## Tools
Capabilities this server exposes over MCP:

- **improve_prompt** — Rewrite a rough prompt into a sharper, ready-to-use prompt structured as Role, Task, Context, Format (RTCF). Returns the improved prompt, its four parts, a short list of clarifying questions worth asking the user, and which model tier and effort level suit this kind of work. Use this before answering when the user's request is vague, or when the user asks to improve a prompt.
- **improve_followup** — Write the user's NEXT message for an AI conversation that is already in progress (a follow-up). Use when the user wants to refine, deepen, or continue earlier work in this conversation. Do NOT use it for a brand-new task (use improve_prompt for that). Returns a short, ready-to-send continuation message that anchors to the existing work instead of restarting it. Requires an API key (add ?key=rk_live_... to the connector URL).
- **build_prompt** — Compose a ready-to-use prompt from explicit Role, Task, Context, and Format parts. Use when the user already knows the pieces and wants them woven into one clean prompt.

## Claude Desktop Quick Installation
Remote MCP endpoint (confidence: high). Install path detected from listing signals. Add as a URL/SSE server in your client:

```json
"mcpServers": {
  "rtcf": {
    "url": "https://rtcf.happlen.com/api/mcp"
  }
}
```

## Documentation

## What the rtcf MCP server does

The rtcf MCP server helps AI users convert underspecified requests into prompts organized around four parts: Role, Task, Context, and Format. It supports Thai and English input, making it suitable for multilingual prompt workflows.

The service exposes two MCP tools. `improve_prompt` takes a rough request, restructures it into the RTCF pattern, and returns a prompt ready for use. `build_prompt` accepts existing Role, Task, Context, and Format material and combines those pieces into a single clean prompt.

## How it works

rtcf runs as a hosted remote MCP server rather than a package that must be installed locally. MCP clients connect to the Streamable HTTP endpoint at `https://rtcf.happlen.com/api/mcp`. After the connection is enabled in a client, an agent can call the prompt tools during a conversation.

The improvement workflow is intended for incomplete or messy requests. The assembly workflow is for cases where the four prompt components already exist but need to be unified. The resulting text can then be used as the prompt for a separate task.

## Setup and configuration

No local installation or OAuth configuration is required for the hosted endpoint. Add the endpoint as a custom MCP connector in a supported client. Claude’s setup uses Settings, Connectors, and Add custom connector; the OAuth fields can remain empty. ChatGPT users on paid plans can enable Developer mode under Settings, Apps, and Advanced before adding the same URL. Claude Code can register it with:

```bash
claude mcp add --transport http rtcf https://rtcf.happlen.com/api/mcp
```

The free tier provides five credits per day without sign-up. `improve_prompt` consumes at least one credit per call, while `build_prompt` is free. Paid-plan users can append a key to the MCP URL, such as `?key=rk_live_...`.

## Tools and capabilities

- `improve_prompt`: Rewrites a rough request into a complete RTCF prompt.
- `build_prompt`: Combines supplied Role, Task, Context, and Format sections.
- Thai and English prompt processing.
- Streamable HTTP access for remote MCP connections.
- REST conversion endpoint at `https://rtcf.happlen.com/api/v1/convert`, authenticated with a bearer key.

The REST endpoint accepts a JSON body containing a `prompt` field and is documented separately by the project. It provides an alternative to invoking the tools through an MCP client.

## Limitations and notes

The rtcf MCP server is a hosted service, so using it requires access to its remote endpoint. The free allowance is limited to five credits per day, and prompt improvement uses credits.

_Full upstream README: https://allmcps.com/mcp/rtcf/readme_