iflow-mcp_shashwatpog_poke-canvas-mcp

# Canvas MCP for Poke Custom MCP server that uses Canvas LMS APIs to provide grades, deadlines, assignments, etc context to Poke What is [Poke](https://poke.com/)? - AI assistant that I use day to day for managing daily tasks, emails, to do lists, etc. via texting on imessage ## Overview - **Authentication**: Canvas Personal Access Token - **Data Access**: Read only, Canvas student account ## Questions this MCP can help answer with Poke - What are my upcoming deadlines? - What announcements were posted recently? - What does my upcoming week look like? - What does my academic day today look like? - Did any of my assignments get graded recently? ## Setting Up Canvas Access Token 1. Log in to your canvas account 2. Go to your Account, Settings Page 3. Scroll to Approved Integrations 4. Create New Access Token (copy the token as you won't see it again) ## Quick Start ```bash git clone https://github.com/Shashwatpog/poke-canvas-mcp cd poke-canvas-mcp # Create a virtual environment and install required packages python -m venv venv source venv/bin/activate pip install -r requirements.txt cp .env.example .env # Edit .env with your Canvas credentials # Run python src/server.py # In a different terminal run the inspector for testing npx @modelcontextprotocol/inspector ``` Server runs on http://localhost:8000/mcp, connect using "Streamable HTTP" Transport ## Environment Variables ```bash CANVAS_BASE_URL=https://your_school.instructure.com CANVAS_ACCESS_TOKEN=you_canvas_access_token ``` ## Deploying The [render.yaml](render.yaml) file contains configuration to deploy on render 1. Fork this repository 2. Create a new web service on render 3. Connect your forked repository 4. Enter the environment variables and deploy ## Poke Integration After deploying the MCP, add the MCP URL in poke settings at [poke.com/settings/connections](https://poke.com/settings/connections) To test it out, ask poke "What do I need to do today on canvas?" "What are my upcoming …