Skip to content

Security: mario/drey

Security

SECURITY.md

Security policy

Reporting a vulnerability

Email mario@djanic.com. Do not open a public issue.

Include what you did, what happened, and how bad you think it is. A proof of concept helps but is not required to start the conversation.

No response-time promise. This is a side project maintained by one person, and an SLA I cannot keep is worth less than an honest "I read everything and I get to it when I can." In practice that is usually days. If a week goes by with nothing from me, ping again; the mail probably got buried rather than ignored.

Coordinated disclosure, and credit in the release notes unless you would rather not have it. If you have waited a reasonable time and want to publish, tell me your date rather than asking permission. Sitting on a real bug indefinitely because a maintainer went quiet helps nobody.

What is in scope

drey brokers connections between LSP clients and a shared language server on one machine, for one user. The interesting failure is cross-client leakage: two clients sharing a server when the sharing rules say they must not, one client seeing another's document contents, or a response routed to the wrong client. If you can make that happen, that is a real bug and I want to hear about it.

Also in scope: the daemon socket's permissions and ownership, anything that lets another user on the same machine attach to your daemon, and the shell wrappers that scripts/install.sh writes onto your PATH.

What is not

drey runs language servers, and a language server executes project configuration. Opening a hostile repository can run code regardless of drey. That is the language server's trust model and drey does not change it.

The daemon is not a network service, does not authenticate, and is not meant to be exposed beyond the local user. If you make it listen on a network, you are outside the design.

Supported versions

Pre-1.0. Fixes land on main and in the next release. There are no backported patch branches yet.

There aren't any published security advisories