Brush Trace (笔痕) - AI Novel Illustration Generator for Openclaw

Brush Trace is an advanced AI assistant skill that scans your novel project's characters and style to generate highly structured, emotionally resonant illustration prompts.

fslong520
v1.3.0
Jun 9, 2026
0
525
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install brushtrace

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install brushtrace using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Brush Trace (笔痕) - AI Novel Illustration Generator?

Brush Trace (笔痕) is a specialized tool designed to solve the challenge of generating highly consistent, style-locked, and emotionally accurate illustrations for web novels and visual stories. As a standout capability in the library of Openclaw Skills, it acts as an intelligent intermediary between your written chapters and image generation models like Midjourney or Stable Diffusion. By reading existing character profiles, historical illustrations, and chapter manuscripts, Brush Trace automatically establishes a visual context, ensuring that characters maintain their unique physical attributes across different scenes.

Unlike generic prompt generators, Brush Trace analyzes the emotional undertone (tone) of each chapter separately from its physical lighting, allowing for rich, dramatic storytelling. Whether it is a warm slice-of-life scene occurring at midnight or a cold, tragic confrontation at high noon, the skill carefully aligns composition, color palettes, and optional comic text layers (designed without invasive speech bubbles or frames) to match the chapter's core narrative.

Brush Trace (笔痕) - AI Novel Illustration Generator Use Cases

  • Maintaining Character Consistency: Instantly scan local markdown files or historical illustration records to ensure character faces, ages, hairstyles, and costumes remain identical across chapters.
  • Chapter Key Scene Extraction: Automatically identify and score high-impact moments (emotional climaxes, action sequences, or quiet atmospheric beats) in a text file to determine which scenes deserve illustration.
  • Generating Production-Ready Prompts: Output structured prompts in Markdown (for in-file notes), JSON (for automated workflows), and clean Midjourney-compatible English text.
  • Embedded Graphic Storytelling: Seamlessly introduce ambient sound effects or narrator text layers that merge directly into the image's texture without artificial white dialogue boxes.

How Brush Trace (笔痕) - AI Novel Illustration Generator Works

  1. Load Context & Build Character Profiles: The agent scans the project directory using glob/grep tools to locate character descriptions and historical design sheets.
  2. Analyze Emotional Tone: It reads the raw chapter text to extract the core emotional undertone (e.g., sweet, suspenseful, warm, tragic), establishing the creative direction before any visual components are chosen.
  3. Extract Scenic Illustration Nodes: The engine scores candidate scenes based on narrative impact and extracts 1 to 3 pivotal moments.
  4. Apply Consistency Controls: Character details are locked using the oldest recorded illustration profiles, preventing random visual drifts.
  5. Synthesize Structured Prompts: The skill compiles a 12-dimensional output (covering subject, moment, characters, environment, lighting, dynamics, composition, color, style, mood, negative prompts, and text layers) into English and Chinese formats.

Brush Trace (笔痕) - AI Novel Illustration Generator Setup

Make sure your workspace environment has appropriate access to filesystem tools. To invoke the brush-trace skill inside an agent session, configure your skill schema with the permitted tools:

# Ensure tools like Read, Write, Glob, Grep, and Task are configured in your workspace
openclaw skill install brush-trace

Create a characters.md or a structured folder with your character sheets so the agent can discover profiles automatically using Openclaw Skills search protocols.

Brush Trace (笔痕) - AI Novel Illustration Generator Data Schema & Taxonomy

The skill generates and structures prompt metadata using a standardized JSON or Markdown taxonomy. This taxonomy ensures your novel's aesthetic remains consistent across generations:

Dimension Field Type Description
tone String Core emotional undertone of the scene (e.g., tragedy, suspense, romantic).
subject String A single-sentence description summarizing the action occurring in the image.
moment String The exact frozen millisecond selected for visual capture.
characters Array of Objects Specific visual attributes of each character involved (id, age, face, pose, hair, costume).
environment Object Location and detail list defining the backdrop constraints.
lighting Object Light direction, temperature, and atmospheric shadows matching the tone.
text_layer Array of Objects Optional, ambient narrative texts (type, position, content, font, effect).
aspect_ratio String Crop configuration mapping the intended distribution medium (e.g., 16:9, 3:4).

Brush Trace (笔痕) - AI Novel Illustration Generator Advanced Features

  • Three-Tier Aesthetics Framework: The generator forces the output through three progressive aesthetic filters: Visual Completeness (Eye-pleasing), Emotional Resonance (Mind-pleasing), and Narrative Metaphor (Spirit-pleasing).
  • Integrated Ambient Text Injection: Supports direct text rendering prompts (such as wind rustling or critical dialogue) designed to emerge natively like ink or frost, maintaining full aesthetic integration without awkward dialog borders.
  • Multi-Output Format Translation: Simultaneously produces structured JSON, structured Markdown, copy-ready Chinese prompt strings, and high-performance Midjourney English prompts.
  • Aspect Ratio Smart Mapping: Dynamically selects aspect ratios based on scene types, prioritizing 16:9 for action, 3:4 for character close-ups, and 2:3 for cover illustration posters.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*