Skip to main contentByteRover offers users a remote space where they can store content from their local space (playbook). This is where you and your team keep all the context you consider necessary and vital for the agentic development process. We’ll help you manage all the context you and your team push to this space.
How context is organized
When local context is pushed to the remote space, ByteRover’s core AI engine performs two key tasks:
- Organizes the context into a logical structure with concepts and their relationships
- Creates vector embeddings for semantic annotation of the concepts
Why context organization matters
You and your team can quickly see how a repository’s context is organized, giving you an instant overview and the ability to make edits as needed.
When an agent retrieves context, the results are accurate and token-efficient.
After reviewing and approving the context (link to brv status) in your local space, push it to the remote space with this command:
Managing context in the remote space
After pushing your context to the remote space, you can access and organize it directly. The remote space provides tools to:
- Review the structured context and concept relationships generated by ByteRover’s AI engine
- Edit and refine the organization to better fit your team’s needs
- Add additional annotations or metadata to enhance context retrieval
- Remove or merge duplicate or redundant concepts
This hands-on management ensures your remote context space remains clean, well-organized, and optimized for agent retrieval. 
