Summary

This project offers an "Agent Skill" functioning as a "visual director" for Xiaohongshu (XHS) image-text posts. It tackles common creator challenges: selecting appropriate visual styles, avoiding generic AI aesthetics, enhancing engagement, and structuring content effectively. The skill benefits users aiming to elevate XHS content, transforming drafts and raw materials into polished, high-impact visual narratives.

How It Works

The XHS Visual Director acts as an AI creative director, analyzing input to recommend optimal visual styles with justifications. A key feature is its two-stage image generation: it produces a confirmation image for user approval on style, composition, and color before generating the full set of 6-8 final images. Alongside visuals, it outputs a complete post structure, per-page plans, AI prompts, publishable copy, and a quality self-check list.

Quick Start & Requirements

Presented as a component within a "Codex Skill" framework, core files reside in the skill/ directory. Specific installation commands, Python versions, or hardware dependencies (GPU/CUDA) are not detailed in the README. Integration into an existing Codex Skill environment is implied.

Highlighted Details

• Socratic Questioning: Initiates detailed dialogues (10-20 questions) for insufficient input to clarify objectives and constraints.
• Dual-Stage Image Generation: A confirmation image allows user feedback before full visual set generation, optimizing resource use.
• Comprehensive Output: Delivers images, structured content plans, visual breakdowns, AI prompts, copy, and a quality checklist for a complete workflow.
• Extensible Style System: Includes mechanisms and templates for defining and adding custom visual styles.

Maintenance & Community

The provided README lacks information on contributors, sponsorships, community channels, or a public roadmap.

Licensing & Compatibility

The README does not specify a software license, presenting a significant adoption blocker due to unclear terms for usage, modification, and distribution, potentially restricting commercial use.

Limitations & Caveats

Dependent on integration within a "Codex Skill" environment. Potential instability in AI-generated Chinese text may require manual intervention. The absence of a specified license is a critical caveat.