ByteRover CLI
The brv CLI memory layer for coding agents: structured project context, git-like versioning and an MCP server
Medium risk
We rate an entry medium when the tool runs code, makes network calls or reads project files. Check what exactly it does before installing.
Why this level
- The CLI reads and writes project files and executes code at the agent's request
- With cloud enabled, some data is sent to ByteRover servers
Install
In your terminal, with SkillFoxx CLI
npx skillfoxx add cli/byterover-cliDetects the agents on your machine, checks the risk and pins the version.
Other ways to install
Install the tool
npm install -g byterover-cliInstall the CLI with npm install -g byterover-cli (or curl -fsSL https://byterover.dev/install.sh | sh on macOS and Linux). Run brv in the project folder for the interactive REPL, or brv mcp to start the MCP server for an agent.
Other ways from the author
curl -fsSL https://byterover.dev/install.sh | shmacOS and Linux, no Node.js required.
This is third-party code. Review the repository files before installing.
What it does
ByteRover CLI gives the brv command an interactive REPL for accumulating project knowledge: architecture facts, decisions, references to specific files. Context is stored as a tree with git-like versioning: branches, commits, merges, push and pull. The tool works with more than 20 LLM providers and ships an MCP server through which any supported coding agent can read and add to this memory. It includes a web dashboard for browsing history and optional cloud sync across machines and teammates.
Who it is for. For developers who want a coding agent to remember project context across sessions and share it with a team.
Good fit when
- The agent should remember project decisions and agreements across sessions
- You need to share accumulated context with other team members
- You need a standalone MCP memory layer that several agents can connect to
Not a fit when
- In-session memory from the agent itself is already enough
- You do not want to run a separate daemon process and context store
Example request
Save to project context that auth uses JWT with a 24-hour tokenLimitations
Some features, such as cross-machine cloud sync, team spaces and the built-in hosted model, require the paid ByteRover Cloud; local use is free. The local REPL needs an API key for the chosen LLM provider.
How to disable. Remove the package with npm uninstall -g byterover-cli, or delete the binary installed by the install script, stopping the background daemon first with brv restart.
Security check
- The CLI reads and writes project files and executes code at the agent's request
- With cloud enabled, some data is sent to ByteRover servers
README in short
The README describes ByteRover CLI as an interactive REPL tool that gives coding agents structured project memory: the agent reads and writes files, executes code, and stores knowledge for future sessions through curate and query commands. It supports more than 20 LLM providers and over 20 coding agents, including Cursor, Claude Code, Windsurf and Cline. The author publishes results on the LoCoMo and LongMemEval-S long-term memory benchmarks. Elastic 2.0 licensed.
FAQ
Is ByteRover Cloud required?
No, everything works locally by default; the cloud adds sync and collaboration on top of local mode.
How does an agent access the saved context?
Through the built-in MCP server started with brv mcp, or directly through the REPL and the brv curate and brv query commands.
Related
Open-source personal AI assistant on your own machine: answers in Telegram, Slack, Discord and WhatsApp, extended with skills and plugins
A self-improving agent from Nous Research with a TUI, messaging gateway, cron jobs and skills it writes itself
An open source coding agent for the terminal and desktop with build and plan modes
Open prompt library with a Claude Code plugin, MCP server and CLI: search, fetch and improve prompts and skills from an agent