| name | content-refactoring-engine |
| description | Automated One-Shot Content Refactoring Engine for NotebookLM drops. Scans incoming drafts in inbox/, generates 2 visual images with captions/alt tags, refactors content for SEO into outbox/master/, and formats channel-specific persona posts for Reddit, LinkedIn, X, and Meta ready for Composio distribution. |
Content Refactoring & Multi-Channel Distribution Engine
This skill automates the complete pipeline from raw NotebookLM drops in inbox/ to SEO-refactored master drafts and multi-channel outbox posts (Reddit, LinkedIn, X, Meta) ready for Composio distribution.
Workflow Execution Steps
When a new drop arrives in inbox/ (or when triggered via "refactor content [file]"):
Step 1: Run SEO Refactor & Image Anchor Selection
Run the SEO refactoring engine script:
python sccd/seo_refactor.py inbox/<drop-filename>.md
This parses the input, generates frontmatter/meta description, scores SEO readability, selects 2 image insertion points, and outputs:
- Refactored master article:
outbox/master/<name>_refactored.md
- Image metadata:
outbox/master/<name>_meta.json
Step 2: Generate 2 Visual Assets
Read the image prompts from outbox/master/<name>_meta.json and invoke the image generator tool (generate_image) for both anchors:
- Image 1: Saved to
outbox/assets/image_1_<slug>.png
- Image 2: Saved to
outbox/assets/image_2_<slug>.png
Attach alt tags, captions, and descriptions in the master markdown file.
Step 3: Transform into Channel Persona Outboxes
Read the refactored master article and apply the persona transformation specifications:
