Update refresh-persona slash command handling (#5)

* Updates related to router slash commands

* lint

* bump jupyter_ai_router dependency version floor

---------

Co-authored-by: David L. Qiu <[email protected]>

Author: 3coins

CLAUDE.md

@@ -5,16 +5,19 @@ This file provides guidance to Claude Code (claude.ai/code) when working with co
 ## Project Overview
 
 This is a JupyterLab extension package that provides default chat commands for Jupyter AI. It consists of:
+
 - A Python server extension (`jupyter_ai_chat_commands`)
 - A TypeScript frontend extension (`@jupyter-ai/chat-commands`)
 
 The extension currently provides two chat commands:
+
 - `@file:<path>`: Add a file as an attachment to a message
 - `/refresh-personas`: Reload local personas defined in `.jupyter/personas`
 
 ## Development Setup
 
 Initial setup requires micromamba/conda and Node.js 22:
+
 ```bash
 micromamba install uv jupyterlab nodejs=22
 jlpm
@@ -26,51 +29,60 @@ The `dev:install` script handles the complete development setup including buildi
 ## Common Commands
 
 ### Building
+
 - `jlpm build` - Build both TypeScript and labextension for development
 - `jlpm build:prod` - Production build with optimization
 - `jlpm build:lib` - Build TypeScript source with source maps
 - `jlpm build:labextension` - Build the JupyterLab extension
 
 ### Development Workflow
+
 - `jlpm watch` - Watch mode for development (runs both TypeScript and labextension watch)
 - `jupyter lab` - Start JupyterLab (run in separate terminal alongside watch)
 
 ### Linting and Testing
+
 - `jlpm lint` - Run all linters (stylelint, prettier, eslint)
 - `jlpm lint:check` - Check without fixing
 - `jlpm test` - Run Jest tests with coverage
 - `pytest -vv -r ap --cov jupyter_ai_chat_commands` - Run Python server tests
 
 ### Extension Management
+
 - `jlpm dev:uninstall` - Remove development extension
 - `jupyter server extension list` - Check server extension status
 - `jupyter labextension list` - Check frontend extension status
 
 ## Architecture
 
 ### Frontend (TypeScript)
+
 - Entry point: `src/index.ts` - exports main plugin and chat command plugins
 - Chat commands: `src/chat-command-plugins/` contains individual command implementations
 - Uses JupyterLab 4.x plugin system and `@jupyter/chat` for chat integration
 
 ### Backend (Python)
+
 - Server extension: `jupyter_ai_chat_commands/extension_app.py`
 - Request handlers: `jupyter_ai_chat_commands/handlers.py`
 - Depends on `jupyterlab_chat`, `jupyter_ai_router`, and `jupyter_ai_persona_manager`
 
 ### Key Dependencies
+
 - Frontend: `@jupyter/chat`, `@jupyterlab/application`, Material-UI icons
 - Backend: `jupyter_server`, `jupyterlab_chat`, `jupyter_ai_router`
 
 ## Code Style
+
 - TypeScript: ESLint with TypeScript rules, Prettier formatting, single quotes
 - Interface naming: Must start with 'I' and use PascalCase
 - CSS: Stylelint with standard config
 - Python: No specific linter configured (follows standard Python conventions)
 
 ## Testing
+
 - Frontend: Jest with coverage reporting
 - Backend: Pytest with pytest-asyncio for server testing
 - Integration: Playwright tests in `ui-tests/` (currently skipped)
 
-Note: Integration tests are currently disabled - see recent commit "skip integration tests for now".
+Note: Integration tests are currently disabled - see recent commit "skip integration tests for now".

jupyter_ai_chat_commands/extension_app.py

@@ -40,21 +40,11 @@ async def initialize_async(self):
         router.observe_chat_init(lambda room_id, ychat: self.on_chat_init(router, room_id))
 
     def on_chat_init(self, router: MessageRouter, room_id: str):
-        router.observe_slash_cmd_msg(room_id, self.on_slash_command)
+        router.observe_slash_cmd_msg(room_id, "refresh-personas", self.on_slash_command)
         self.log.info("Attached router observer.")
 
-    def on_slash_command(self, room_id: str, message: Message):
-        first_word = get_first_word(message.body)
-        assert first_word and first_word.startswith("/")
-
-        command_id = first_word[1:]
-        if command_id == "refresh-personas":
-            self.event_loop.create_task(self.handle_refresh_personas(room_id))
-            return True
-
-        # If command is unrecognized, log an error
-        self.log.warning(f"Unrecognized slash command: '/{command_id}'")
-        return False
+    def on_slash_command(self, room_id: str, command: str, message: Message):
+        self.event_loop.create_task(self.handle_refresh_personas(room_id))
 
     async def handle_refresh_personas(self, room_id: str):
         self.log.info(f"Received /refresh-personas in room '{room_id}'.")

pyproject.toml

@@ -28,7 +28,7 @@ classifiers = [
 dependencies = [
     "jupyter_server>=2.4.0,<3",
     "jupyterlab_chat>=0.17.0,<0.18.0",
-    "jupyter_ai_router>=0.0.1",
+    "jupyter_ai_router>=0.0.2",
     "jupyter_ai_persona_manager>=0.0.3",
 ]
 dynamic = ["version", "description", "authors", "urls", "keywords"]