Author Style

Provides deterministic author-style coordinates and prompts for text and image generation based on a catalog of 11 styles across 8 dimensions.

python

0

GitHub Stars

python

Language

4 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "dmarsters-author-style-mcp": {
      "command": "python",
      "args": [
        "author_style_mcp.py"
      ]
    }
  }
}

You set up and use a modular MCP server that provides deterministic, high-dimensional author style coordinates to drive text and image prompts. You can blend styles, measure distances between styles, and generate structured prompts for downstream LLMs and image generators. This enables consistent, controllable creative transformations across text and visuals while keeping the core processing stateless and deterministic.

How to use

You interact with the MCP server by running the local Python service and then connecting your MCP client to it. The server exposes deterministic style tools that let you query author style coordinates, blend styles, and request text or image prompts tailored to a chosen style or blend.

How to install

Prerequisites: you need Python installed on your system. Ensure you have Python 3.8+ available on your path.

Step 1: Clone the repository containing the MCP server code.

Step 2: Navigate to the server directory and install dependencies.

Step 3: Start the local MCP server.

Additional sections

Configuration notes: the server is designed to run as a local process and expose a standard input/output interface for integration with an MCP client. Use the provided command to start the server and establish a connection from your client.

Example workflow: obtain coordinates for Hemingway, blend with Lovecraft and Shōnagon, and generate a text prompt that encodes the blended style. You can also request image prompts with a style modifier to guide visual generation.

Troubleshooting tips: ensure Python is on your PATH, check that the working directory contains the server script, and verify that your client connects using the expected I/O channel. If the server fails to start, review dependency installation and Python version compatibility.

Available tools

get_author_styles

List all 11 author styles with their coordinates and signatures.

get_author_style_profile

Return a complete profile for a given style, including coordinates, signature moves, and vocabularies for text and image generation.

get_style_dimensions

Provide all 8 style dimensions with their low, mid, and high mappings for both text and image outputs.

get_parameter_names

Return the ordered parameter list used for dynamics integration and style morphing.

compute_author_distance

Calculate Euclidean distance between two author styles and provide a per-dimension breakdown.

blend_author_styles

Interpolate multiple styles to produce a blended coordinate set and combine vocabularies.

generate_text_style_prompt

Produce structured directives for text generation based on a specific style, blend, or custom coordinates.

generate_image_style_prompt

Produce visual vocabulary, color palettes, and compositional rules for image generation based on a style or blend.

find_style_extremes

Identify the maximum-contrast pair across the catalog.

find_nearest_style

Find the closest neighboring style in style-space to a given style.