Session Relay runs on your own machine — the client starts it, so there is no endpoint to ping. Last commit 24 Oct 2026.
Share coding-agent sessions with your team: /share in one agent, /pickup in another.
We read the source, 3 h ago · rules 995398eba69f
What this server is able to do. For an MCP server this is often the job itself — a terminal server runs commands because that is what it is for. Listed so you know what you are plugging in, not as an accusation.
spawn(cmd, args, { stdio: "ignore", detached: true }).on("error", () => undefined).unref();
${file ? sql`AND EXISTS (SELECT 1 FROM unnest(s.files_touched) f WHERE f ILIKE ${"%" + file.replace(/[\\%_]/g, (ch) => "\\" + ch) + "%"})` : sql``}
Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.
We found places where it runs commands, builds paths or queries from values it is given. None of that is a flaw by itself — it becomes one when the code changes, and code changes quietly between releases. We re-read it on every one.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add session-relay -- npx -y session-relay
{
"mcpServers": {
"session-relay": {
"command": "npx",
"args": [
"-y",
"session-relay"
]
}
}
}
[mcp_servers.session-relay]
command = "npx"
args = ["-y", "session-relay"]
{
"mcpServers": {
"session-relay": {
"command": "npx",
"args": [
"-y",
"session-relay"
]
}
}
}
{
"servers": {
"session-relay": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"session-relay"
]
}
}
}
Optional settings from the registry entry, left out of the block above:
RELAY_SERVER (URL of your team's Session Relay server. Optional if you ran relay login), RELAY_TOKEN (Personal access token (srk_...) from Settings → Tokens. Optional if you ran relay login).
An AI agent reads your server over the one SSH session you are already watching.
Mine your AI coding session logs into a you.md profile with dated receipts, for an agent to read.
Shared memory for coding agents. Stop re-explaining your codebase every session.
The team layer for AI coding agents: shared contracts, collision alerts, E2EE sessions.
Share a live agent coding session by URL. Spectate or invite in.
Coding agents in one repo see each other, get warned before shared edits, and message.
One shared brain for your AI coding agents: team memory, agent Q&A, tasks, and file claims.
Local mail between coding-agent sessions on one machine; relays to a running Swarmail server.
Answers built from our own checks of this server.