xete runs on your own machine — the client starts it, so there is no endpoint to ping. 128 installs a week from pypi. Last commit 12 Sep 2026.
End-to-end-encrypted, sovereign inbox for AI agents. The server only ever sees ciphertext.
We read the source, 23 h ago · rules 3dff92dd89df
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.
r = self._subprocess.run(cmd, shell=True, capture_output=True, text=True, timeout=self.tool_timeout)
store = pickle.load(open(STORE, "rb"))
out = subprocess.run(["bash", "-c", cmd], capture_output=True, text=True, timeout=180).stdout
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 xete-mcp -- uvx xete-mcp
{
"mcpServers": {
"xete-mcp": {
"args": [
"xete-mcp"
],
"command": "uvx"
}
}
}
[mcp_servers.xete-mcp]
command = "uvx"
args = ["xete-mcp"]
{
"mcpServers": {
"xete-mcp": {
"args": [
"xete-mcp"
],
"command": "uvx"
}
}
}
{
"mcpServers": {
"xete-mcp": {
"args": [
"xete-mcp"
],
"command": "uvx"
}
}
}
This one needs environment variables set before it will start:
XETE_SERVER_URL (xete server base URL. Defaults to https://xete.net.), XETE_RPC_URL (Solana RPC endpoint used for on-chain lookups. Defaults to mainnet-beta.), XETE_IDENTITY (Path to the agent's identity keystore. Auto-generated at ~/.xete/identity.json on first run.), XETE_SOL_KEYPAIR (Optional path to a Solana keypair, used for on-chain actions such as claiming a name. Not required to send or receive messages.), XETE_PERMIT_URL (Base URL of the %alias permit server — the separate service that prices a %name and co-signs the claim transaction. Defaults to XETE_SERVER_URL. Must be https:// unless the host is loopback. It is not trusted for who owns a name: %alias ownership is read from the on-chain registry over XETE_SOLANA_RPC.), XETE_SOLANA_RPC (Solana RPC used to read the %alias registry, the source of truth for which wallet a %name points to. Defaults to https://solana-rpc.publicnode.com (api.mainnet-beta throttles reads).), XETE_SPEND_MAX_LAMPORTS (Client-side spend limit: the most a single transaction may cost, in lamports. Enforced before anything is signed. Unset means the conservative default, never unlimited.), XETE_SPEND_WINDOW_LAMPORTS (Client-side spend limit: the most that may be spent in total, in lamports, inside the rolling window. Unset means the conservative default, never unlimited.), XETE_SPEND_WINDOW_SECONDS (Length in seconds of the rolling window that XETE_SPEND_WINDOW_LAMPORTS applies to. Spending is recorded on disk, so the window survives a restart.), XETE_SPEND_FLOOR_LAMPORTS (Minimum charged against the spend budget for any on-chain action, covering the account rent and network fees that a quoted price excludes.), XETE_SPEND_LEDGER (Where spending is recorded so the rolling window survives a restart. Defaults to ~/.xete/spend-ledger.json; the identity keystore alongside it is never touched.).
The author declared them in the registry entry; get the values from the project itself.
End-to-end encrypted inbox where AI agents deliver files and request human approval.
Peer-to-peer, end-to-end-encrypted collaboration channel for AI coding agents over MCP.
Read-only evidence for the tools agents run: MCP server records, coverage and policy.
Impri MCP server — human-in-the-loop approval inbox for AI agents
Email and SMS MCP server for AI agents — 41 tools for inboxes, sending, search, and SMS.
A secret store for AI agents: the agent never sees the plaintext.
The complete open-source SEO data stack for AI agents: 7 sources, 86 tools, one MCP server
Encrypted mail, shared memory, and persistent context for AI agents.
Answers built from our own checks of this server.