ByteRover CLI Freemium

What it does

ByteRover CLI (brv) gives AI coding agents persistent, structured memory. Run brv in a project and it opens an interactive REPL, powered by an LLM of your choice, that curates project knowledge into a "context tree," lets you query it, and syncs it to the cloud so the same knowledge can be shared across tools and teammates.

What is inside

• 24 built-in agent tools spanning file ops, bash execution, knowledge create/expand/search, and memory read/write/edit/delete
• Git-like version control for the context tree: brv vc branch/commit/merge/push/pull
• A review workflow (brv review pending/approve/reject) for human-in-the-loop approval of curated changes
• A built-in MCP server (brv mcp) so other agents can query/curate its memory over the Model Context Protocol
• A Hub/connectors ecosystem for installable skills and bundles

Works with

The README states support for 22+ AI coding agents (Cursor, Claude Code, Windsurf, Cline, and more) and 20 LLM providers (Anthropic, OpenAI, Google, Groq, Mistral, xAI, DeepSeek, and others).

How to install or connect

# Shell script (macOS & Linux, no Node.js required)
curl -fsSL https://byterover.dev/install.sh | sh

# npm (all platforms, requires Node.js >= 20)
npm install -g byterover-cli

# Verify
brv --version

First run: cd your/project && brv — the REPL auto-configures with no setup needed.

Maintenance and safety

Maintained by the ByteRover team with a public Discord; last commit to the repo was about three months before this review. It ships under the Elastic License 2.0, a source-available license, not a fully open-source one (it restricts offering the software as a competing hosted service). The core CLI runs locally; ByteRover Cloud adds sync, a hosted LLM, and team management as a separate, more commercial layer.

Who should use it

Developers already using multiple AI coding agents (Cursor, Claude Code, Windsurf, etc.) who want a shared, version-controlled memory layer between them, especially teams that want reviewable, auditable changes to that shared context rather than an agent silently rewriting its own notes.