Skip to main content
This tutorial demonstrates core features of ByteRover:
  • Intelligent local curation with brv curate (executed by your coding agent via your prompts)
  • Push organized context to remote space
  • Agentic search with brv query (executed by your coding agent via your prompt)

​
Start with Your Local Context

We start by creating a local context file named security.md. In a real scenario, you could already have many of these files in your codebase. This is the content of the security.md file: 
Security Best Practices:
Never commit .env files - Add to .gitignore
Use strong secrets - Minimum 32 characters for JWT secrets
Rotate secrets regularly - Especially in production
Use secret management - AWS Secrets Manager, HashiCorp Vault
Validate on startup - Fail fast if configuration is wrong

​
Curate to Your Context Tree

Copy this prompt and paste it into your coding agent’s chat (e.g., Cursor, Claude Code, etc.): 
> use brv curate to intelligently organize content from security.md
Your coding agent will execute the command: 
brv curate "Security Best Practices:
Never commit .env files - Add to .gitignore
Use strong secrets - Minimum 32 characters for JWT secrets
Rotate secrets regularly - Especially in production
Use secret management - AWS Secrets Manager, HashiCorp Vault
Validate on startup - Fail fast if configuration is wrong"
ByteRover automatically analyzes the content and organizes it into the context tree, detecting relevant domains (likely compliance or code_style) and creating an appropriate topic structure.

​
Review Your Changes

Before pushing, check what was added: 
# In the REPL:
/status
This shows all added topics and subtopics in your context tree, allowing you to review the organization before syncing to remote.

​
Push to Remote Context Space

From the terminal, run: 
# In the REPL:
/push
Check the ByteRover web application to confirm the context has been stored—this process should be very fast. Copy this prompt and paste it into your coding agent’s chat: 
> query the context tree for security best practices using brv query
ByteRover will search intelligently, follow relations between topics, and synthesize an answer from your context tree. You can even run the query yourself to curate the context for your agent more effectively.

​
What just happened?

Step 1: Agent Executes Curate Command Your coding agent read security.md and executed brv curate with the content. ByteRover then processed it for intelligent organization. Step 2: Intelligent Curation ByteRover automatically analyzed the content and:
  • Detected relevant domains (compliance/security)
  • Organized content hierarchically into the context tree
  • Created context.md files in .brv/context-tree/
  • Added relations to connect related topics
Step 3: Review Changes Running brv status showed the newly added topics and subtopics, allowing you to verify the organization before pushing. Step 4: Push to Remote Running brv push synced your organized context tree structure to ByteRover’s remote space, making it available to your team. Step 5: Agentic Search When you queried with brv query, ByteRover:
  • Reasoned about what you were asking for
  • Searched the context tree using strategic patterns
  • Followed relations between topics for comprehensive context
  • Synthesized a coherent answer with citations

​
But why does this matter?

ByteRover acts as your intelligent context engineer—it understands meaning, not just keywords. Try one of these next to see the full power of ByteRover:

​
Curate context inline

Instead of starting from a file, copy this prompt and paste it into your coding agent’s chat: 
> use brv curate to add all the context for best practices in development with typescript
ByteRover’s goal is to make context management effortless for you and your team. That means no more “markdown hell” when managing multiple markdown files, no more wasted tokens reading through these files, and no context collisions when you and your teammates work on the same files. Intelligent organization means ByteRover:
  • Detects which domains your knowledge belongs to
  • Prevents duplication by searching existing context
  • Creates hierarchical structure with explicit relations
  • Understands semantic meaning, not just text similarity

​
Curate multiple files

Copy this prompt and paste it into your coding agent’s chat: 
> read src/auth/oauth.ts and src/auth/tokens.ts, then use brv curate to add context about the authentication implementation
Your coding agent will read both files and execute brv curate with their combined context. ByteRover will then intelligently organize everything into the appropriate domains and topics.

​
You can have more interesting workflows

Combine curation and querying to build on team knowledge: Copy this prompt and paste it into your coding agent’s chat: 
> query existing authentication patterns, then curate my new OAuth implementation
Your coding agent will retrieve relevant context using brv query, then add your new knowledge using brv curate with proper organization and relations.