ServeRay FAQs

Question 1

Can ServeRay perform write operations?

Accepted Answer

By default, ServeRay operates in a read-only mode to maximize security. While database write support is available, it requires explicit opt-in with an enable flag and acknowledgement string, and is subject to strict server-side enforcement.

Question 2

What integrations does ServeRay support?

Accepted Answer

ServeRay provides robust integrations for MSSQL, PostgreSQL, MySQL databases, GitHub for issues/PRs, Jira for project tracking, Slack for channel history, local filesystems, and Playwright for constrained browser navigation.

Question 3

Is ServeRay a hosted service or a multi-tenant platform?

Accepted Answer

No, ServeRay is explicitly not a hosted platform, shared control plane, or multi-tenant service. It provides download-and-run binaries for you to install, configure with your own credentials, and operate locally on your infrastructure.

Question 4

What is ServeRay?

Accepted Answer

ServeRay is a collection of local-first, self-hosted Model-Context-Protocol (MCP) servers designed for secure, read-only access to various systems including databases, GitHub, Jira, Slack, filesystems, and Playwright, all run on your own machine.

Question 5

How does ServeRay ensure data security?

Accepted Answer

ServeRay prioritizes security with safe-by-default principles. It enforces read-only access by default, uses server-side allowlists for repositories, projects, channels, and origins, and blocks dangerous operations like arbitrary writes or file traversal.

ServeRay

ServeRay

主な機能

ユースケース

主な機能

ユースケース