xhs-autopilot
Red (Xiaohongshu) Full-Autonomous AI-Native Workflow Alchemy System. 30-min operation loop with self-improvement.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/changer-changer/xhs-autopilotWhat This Skill Does
xhs-autopilot is a high-level, AI-native autonomous workflow system designed specifically for Xiaohongshu (Red) content operations. It operates on a 30-minute closed-loop cycle, integrating advanced 'alchemy' techniques to handle everything from content strategy alignment and trend research to multi-modal creation and automated posting. The system features a sophisticated three-layer memory architecture that isolates technical workspace configurations from content-specific semantic memories and real-time execution logs. It is designed to act as a self-improving entity that reflects on its own output, analyzes engagement metrics via computer vision and sub-agents, and systematically addresses performance bottlenecks.
Installation
To integrate this agentic system into your workspace, run the following command in your terminal: clawhub install openclaw/skills/skills/changer-changer/xhs-autopilot
Ensure your environment has access to Python 3 and the necessary dependencies for Playwright, as the system utilizes Playwright CDP modes for browser interactions. After installation, you can initialize the workflow manually using the execution script located in scripts/autopilot/execute_loop.py or trigger the full background service via run.sh.
Use Cases
- Automated Influencer Operations: Maintain a consistent, high-quality posting frequency without manual intervention.
- Content A/B Testing: Automatically track engagement performance and let the self-improvement loop iterate on title and cover visual strategies.
- Trend Hijacking: Utilize the research phase to identify current platform trends and rapidly synthesize them into content pieces using pre-defined technical and cyberpunk aesthetic constraints.
- Data-Driven Strategy: Continuously refine your target audience and tone based on actual comment analysis and performance feedback loops.
Example Prompts
- "xhs-autopilot, initiate the next 30-minute optimization cycle and report on the performance of the most recent post."
- "Review the current bottleneck in BOTTLENECKS.md and update the creative strategy for the next batch of content."
- "Run a full audit on the last 5 posts to verify if they strictly adhere to our JetBrains Mono font and cyberpunk visual guidelines."
Tips & Limitations
- Strict Memory Management: Never store generic technical configurations in the xhs-memory folders; keep memory layers separated to ensure modularity.
- Visual Consistency: Always ensure your local environment supports the required font assets (JetBrains Mono) to avoid rendering errors in cover generation.
- Autonomy Warning: Because this system performs automated publishing (Act phase), always monitor the first few cycles in a sandbox environment to ensure the AI's 'persona' matches your brand expectations before allowing fully autonomous operation.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-changer-changer-xhs-autopilot": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: network-access, file-write, file-read, data-collection, external-api, code-execution
Related Skills
project-context-manager
Project-based agent context management system for maintaining long-term memory and project state across sessions. Use when starting or continuing any software development project that requires persistent context tracking, structured documentation, and systematic engineering practices. This skill enforces PROJECT_CONTEXT.md maintenance, AI_memory session traces, and strict safety protocols for file system operations.
paper-research-agent
Autonomous multi-agent paper research system. When user wants to research a topic, find related papers, or analyze academic literature, use this skill to orchestrate the full pipeline: intelligent search → PDF download → parallel agent analysis → comprehensive report generation. Triggers on: "research papers on X", "find related literature", "analyze papers", "调研论文", "查找相关文献", "分析论文", "帮我调研XXX领域"
paper-reader
Comprehensive PDF paper reader for academic research. Extracts text, figures, tables, and structured content from research papers with support for multimodal analysis.