Byterovermemory MCP

What byterovermemory-mcp MCP server does

The byterovermemory-mcp MCP server exposes ByteRover Team Memory as a hosted remote service for AI agents and coding teams. It is intended for OpenAI Codex memory workflows and provides structured JSON operations rather than a local database or repository implementation.

Its four available tools are memory_recall, memory_write, memory_conflict_check, and memory_receipt_export. Together, they cover retrieving stored memory, adding memory, checking for conflicts, and exporting receipt information associated with usage or workflow activity.

The service also publishes a public MCP server card and a product website with plan and token information. The repository itself is documentation and directory-submission material; it does not contain the private production source code.

How it works

The byterovermemory-mcp MCP server is accessed through the remote endpoint at https://byterovermemory.clauxel.com/mcp. Clients communicate with it using MCP over Streamable HTTP and send JSON-RPC requests. Production requests must include a bearer token in the HTTP authorization header:

Authorization: Bearer <token>

A token is obtained from the product website after selecting a plan or requesting access. Browser requests made without authentication receive a JSON error rather than internal service details. The service also provides a server-card document at /.well-known/mcp/server-card.json and identifies itself in the registry as com.clauxel.byterovermemory/byterovermemory-mcp.

Setup and configuration

1. Open the ByteRover Team Memory website and select a plan.
2. Create or request an API token.
3. Configure an MCP client that supports remote Streamable HTTP servers with the MCP endpoint.
4. Supply the token as a bearer credential when sending JSON-RPC requests.

No environment variable names or local installation procedure are documented in the supplied material. The service is hosted, so running the repository locally is not presented as an installation path.

Tools and capabilities

• memory_recall retrieves memory through the hosted service.
• memory_write writes memory to the service.
• memory_conflict_check checks memory for conflicts.
• memory_receipt_export exports memory-related receipt information.

The available interface is designed around structured JSON tools. The documentation also identifies usage receipts and audit-ready workflow evidence as part of the hosted service, but does not specify the individual input fields or response schemas for these tools.

Limitations and notes

The service is paid and requires a production bearer token. It is remote-only in the documented setup, so client connectivity and Streamable HTTP support are prerequisites. The supplied documentation does not describe offline use, local persistence, pricing details, token lifetime, rate limits, or detailed tool schemas. It also does not name specific MCP clients; compatibility depends on whether the chosen client supports remote Streamable HTTP servers and bearer authentication.