Prioritize Guides Over Blogs in AI Queries

To mitigate the risk of the AI using outdated information from historical blog posts, a new content-type priority system will be implemented. This ensures that evergreen content (guides, source code) is always favored over time-stamped blog posts for technical and implementation-related queries.

This change follows the "Content-Type Priority" approach brainstormed to improve the long-term integrity of the knowledge base.

Scope of Work

1. Update queryKnowledgeBase.mjs Scoring:

   • Modify the scoring algorithm to significantly increase the score for guide chunks compared to blog chunks. This will ensure that for a given keyword match, a guide will always be ranked higher than a blog post.

2. Update AGENTS.md Protocol:

   • Add a new instruction to the "How to Interpret Query Results" section.
   • This instruction will explicitly state that the agent must prioritize guide and src results for implementation details, and should treat blog posts as sources for historical and conceptual context only.

3. Establish New Convention:

   • This ticket formalizes a new, forward-looking convention: all future blog posts should ideally include a version number in their filename or front matter (e.g., v10.9-My-Feature.md) to provide clear temporal context.

Acceptance Criteria

• A general query that matches keywords in both a guide and a blog post ranks the guide significantly higher.
• AGENTS.md contains the new directive for prioritizing query results.
• The new convention for versioning blog posts is documented.