Core Capabilities

Core Capabilities

GEOFlow now covers a complete operating loop from knowledge assets to AI generation, review, publishing, multi-site distribution, and analytics.

1. Model Management

• OpenAI-compatible providers and native Gemini endpoints
• chat and embedding model types
• provider base URL and full endpoint adaptation
• model connection tests, usage statistics, and troubleshooting signals
• smart failover and priority control
• default embedding model configuration

2. Knowledge Bases and RAG

• knowledge upload, chunking, preview, and vectorization status
• structured rule chunking
• optional LLM semantic chunk planning
• stable fallback when semantic planning fails
• chunk metadata for title, section path, strategy, sequence, and source hash
• task-level knowledge retrieval during article generation

3. Materials and Prompts

• title libraries
• keyword libraries
• image libraries
• author libraries
• body prompts
• special prompts
• material health overview

4. Task Scheduling

• task creation and editing
• generation count, publishing cadence, review mode, and task status
• scheduled dispatch, queues, and workers
• fixed model and smart failover modes
• publication scope: local plus channels, channels only, or local only
• retry handling and execution tracking

5. Content Workflow

• draft generation
• manual review and auto approval
• publishing-state control
• trash and restore handling
• local article and remote-copy edit/delete tracking
• SEO title, description, keywords, category, and author management

6. Frontend Delivery

• homepage, category, archive, and article pages
• SEO metadata, Open Graph, and structured data
• GFM Markdown rendering for headings, lists, tables, code blocks, quotes, and images
• uploaded image path normalization
• theme preview, theme packages, and activation

7. Multi-site Distribution

• GEOFlow Agent target-site packages
• WordPress REST API channels
• encrypted Application Passwords and Agent secrets
• connection tests, remote settings sync, queue, and logs
• static mode and rewrite mode
• remote homepage, article pages, sitemap, llms.txt, and TXT map generation
• remote article editing, deletion, and local back-write where applicable

8. Analytics

• system overview metrics
• single-site content operations
• multi-site distribution analytics
• self-service access-log analysis
• top content, top channels, status codes, source types, and AI crawler recognition
• filters by date range, quick time range, task, channel, category, article, and log source

9. Admin Operations and Extension

• multilingual admin UI
• admin account editing and deletion
• GitHub version update reminders
• first-deployment guide
• web admin, API, CLI, and Skill ecosystem
• template, design, and CLI companion skills

The real value is not that these modules exist separately, but that they form one system path:

knowledge base -> materials -> tasks -> queue -> generation -> review -> publishing -> distribution -> analytics -> ongoing maintenance