ByteRover

# ByteRover Persistent knowledge management for AI agents — store and retrieve project patterns, decisions, and architectural rules across sessions using the brv CLI. ByteRover provides long-term memory for AI agents through a local knowledge management system. Using the brv CLI, your AI agent can store project patterns, architectural decisions, coding rules, and other contextual information as human-readable Markdown files in .brv/context-tree/ . Before any task, the agent queries this knowledge base to recall relevant context, ensuring consistency across sessions. ## How It Works The workflow is simple: before thinking about a task, run brv query to retrieve relevant existing context from the knowledge base. After implementing changes, run brv curate to save new patterns and decisions. Both commands use a configured LLM provider (ByteRover's default provider works with no API key) to synthesize answers from stored context and to categorize new entries. Knowledge is stored as Markdown files that are version-controllable and human-readable. ## Key Features - **Query Knowledge**: Retrieve relevant context before any task using brv query with natural language questions - **Curate Context**: Save new patterns, decisions, and rules with brv curate , optionally including up to 5 source files - **No API Key Required**: The default ByteRover LLM provider works out of the box for query and curate operations - **Multiple LLM Providers**: Optionally connect OpenAI, Anthropic, Google, or other providers with your own API key - **Curate History**: View past curations with brv curate view , filter by time and status - **Optional Cloud Sync**: Push and pull knowledge to ByteRover's cloud for team sharing (requires authentication) - **Human-Readable Storage**: All knowledge stored as Markdown files in .brv/context-tree/ , fully version-controllable ## Requirements - **Node.js**: Required to install the brv CLI via npm ## Use Cases - **Cross-Session Consistency**: Ensure your AI agent remembers project conventions, coding patterns, and past decisions across multiple sessions - **Architectural Documentation**: Store and recall architectural rules, authentication patterns, database schemas, and design decisions - **Team Knowledge Sharing**: Sync curated knowledge across team members via optional cloud sync - **Onboarding Acceleration**: New team members or AI agents can quickly query the knowledge base to understand project conventions ## Installation Install via: `npx clawhub@latest install byterover`

Instalación

Habilidades relacionadas