@mulmobridge/telegram 1.0.2 → 1.1.0

package/README.md

@@ -28,15 +28,17 @@ yarn telegram
 |---|---|---|---|
 | `TELEGRAM_BOT_TOKEN` | Yes | BotFather token | - |
 | `TELEGRAM_ALLOWED_CHAT_IDS` | Yes | Comma-separated integer chat IDs | empty (deny all) |
-| `MULMOCLAUDE_API_URL` | No | Server URL | `http://localhost:3001` |
+| `MULMOCLAUDE_API_URL` | No | Server URL | auto (`.server-port`; waits if nothing is published) |
 | `MULMOCLAUDE_AUTH_TOKEN` | No | Bearer token override | reads from file |
 | `TELEGRAM_POLL_TIMEOUT_SEC` | No | Long-poll timeout | `25` |
 
 ### Auth token persistence across server restarts
 
-The MulmoClaude server regenerates a fresh bearer token on every startup and writes it to `~/mulmoclaude/.session-token`. The bridge reads that file once at launch and keeps the token in memory — so if the server restarts while the bridge is running, the bridge keeps using the **old** token and every API call returns **401**, silently.
+The MulmoClaude server regenerates a fresh bearer token on every startup and writes it to `<workspace>/.session-token` (`$MULMOCLAUDE_WORKSPACE_PATH`, or `~/mulmoclaude` when unset), alongside the port it bound in `.server-port`.
 
-**Fix**: set `MULMOCLAUDE_AUTH_TOKEN` to the same long random value on **both** the server and the bridge. The server uses it verbatim instead of regenerating, so the token survives restarts and the bridge stays authenticated.
+**The bridge follows a restart on its own.** When the connection fails it re-reads both files, and if the server came back as a different generation — new token, new port, or both — it rebuilds its socket against it (#3078). You do not have to restart the bridge.
+
+Pinning the token is still useful when the bridge runs **on a different machine** from the server, where it cannot read the workspace at all: set `MULMOCLAUDE_AUTH_TOKEN` to the same long random value on both sides. The server then uses it verbatim instead of regenerating.
 
 ```bash
 # Server (one-time setup — same value across restarts)

package/dist/index.js

@@ -12,7 +12,7 @@
 //   MULMOCLAUDE_AUTH_TOKEN      — optional override
 //   TELEGRAM_POLL_TIMEOUT_SEC   — optional, default 25
 import "dotenv/config";
-import { createBridgeClient } from "@mulmobridge/client";
+import { createBridgeClient, installProcessGuards } from "@mulmobridge/client";
 import { createTelegramApi } from "./api.js";
 import { parseAllowlist } from "./allowlist.js";
 import { createMessageRouter } from "./router.js";
@@ -88,11 +88,15 @@ async function main() {
         router.handlePush(ev).catch((err) => console.error(`[telegram] handlePush failed: ${String(err)}`));
     });
     const abortController = new AbortController();
-    process.on("SIGINT", () => {
-        console.log("\nShutting down...");
-        abortController.abort();
-        client.close();
-        process.exit(0);
+    // Installed here rather than at module level because the shutdown work needs
+    // the controller and the socket. `main().catch` still covers the awaited
+    // startup path above, so nothing is unguarded in between.
+    installProcessGuards({
+        name: TRANSPORT_ID,
+        onShutdown: () => {
+            abortController.abort();
+            client.close();
+        },
     });
     await pollLoop(api, router, {
         pollTimeoutSec,

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@mulmobridge/telegram",
-  "version": "1.0.2",
+  "version": "1.1.0",
   "description": "Telegram bridge for MulmoBridge — connect a Telegram bot to MulmoClaude",
   "type": "module",
   "main": "./dist/index.js",
@@ -30,12 +30,12 @@
   "license": "MIT",
   "author": "Receptron Team",
   "dependencies": {
-    "@mulmobridge/client": "^1.0.2",
+    "@mulmobridge/client": "^1.2.0",
     "@mulmobridge/protocol": "^1.0.1",
     "dotenv": "^17.0.0"
   },
   "devDependencies": {
-    "@types/node": "^26.1.2",
+    "@types/node": "^26.4.1",
     "typescript": "^6.0.3"
   },
   "homepage": "https://github.com/receptron/mulmoclaude/tree/main/packages/bridges/telegram#readme",