| name | trend-to-redbook-automation |
| description | End-to-end automation orchestrator. Fetches data from Twitter (X) or global News MCPs, passes it through the detoxifier for compliance, formats it via the content-creator, and renders/publishes to Redbook using the local render engine. |
🌟 Trend to Redbook Automation Workflow
Core Definition
This acts as a "Workflow Orchestrator". It does not perform physical work itself, but schedules and coordinates operations across four distinct modules, creating a complete, no-code AI automation pipeline.
Prerequisites
To run this pipeline perfectly, you need to install the following underlying skill modules in OpenClaw/your system:
- Data Acquisition Layer (Install at least one):
opentwitter-mcp (from infra403, for stable Twitter fetching)opennews-mcp (from infra403, for stable news sourcing)
- Processing & Publishing Layer (Required standard suite):
redbook-anti-risk-detoxifier-skills (Compliance formatting & deduplication)redbook-content-creator-skills (Redbook viral copywriting structure)redbook-render-skills (Local image rendering & publishing engine)
Automation Pipeline SOP (Execution Standard for Agents)
When the user issues a command (e.g., "Help me migrate @cryptoxiao's latest tweet to Redbook"), the Agent must strictly and sequentially execute the following four stages in the background. Do not arbitrarily interrupt or ask questions unless an error occurs:
🎯 Stage 1: Information Acquisition
- Call the appropriate MCP interface based on the user's source command.
- If the target is a Twitter influencer (e.g.,
@cryptoxiao), silently call opentwitter-mcp to get their latest tweet (Text/Media).
- If the target is global news, call
opennews-mcp to fetch the news content.
- Output: Obtain high-density
Raw_Text.
🎯 Stage 2: Compliance Detoxification
- Pass the
Raw_Text to redbook-anti-risk-detoxifier-skills.
- Strictly follow its ban-word list and detoxification strategy (e.g., rewriting high-risk terms like "VPN", "Twitter scraping", "Crypto" into localized neutral phrases like "cloud assistants" and "information arbitrage").
- Output: Obtain a sanitized and compliant
Safe_Text.
🎯 Stage 3: Viral Content Creation
- Feed the
Safe_Text into redbook-content-creator-skills.
- Format the content following the CES algorithm rules for strong emotional pacing.
- Output: A structured final draft, including a highly suspenseful
Title, a platform-native Body, and high-traffic Tags.
🎯 Stage 4: Rendering & Publishing
- Locate the root directory of
redbook-render-skills (formerly Auto-Redbook-Skills).
- Inject the title and body generated in the previous step, and call its core
render_xhs.py or render_xhs.js to batch-render stacked graphic cards.
- Finally, execute the publishing script or prompt the user to review the beautifully rendered card set in the local output directory.
Why Design It This Way?
Novice users or operators only need to trigger this skill with a single sentence. The AI will act as the master contractor of the pipeline, automatically dispatching the Twitter data fetcher, the compliance filter, the copywriter, and the graphic renderer, delivering a finished product ready for Redbook publishing within a minute.
