# Kiro CLI Integration Guide
This guide shows how to integrate the Canvas LMS MCP Server with Amazon's Kiro CLI for seamless Canvas access through AI assistance.
## Prerequisites
- **Kiro CLI** installed and configured
- **Canvas API Token** from your institution
- **Python 3.10+** with virtual environment support
## Installation Steps
### 1. Clone and Install the Canvas MCP Server
```bash
# Clone the repository
git clone https://github.com/sweeden-ttu/canvas-lms-mcp.git
cd canvas-lms-mcp
# Create and activate virtual environment
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
# Install dependencies
pip install -e .
```
### 2. Configure Canvas Credentials
```bash
# Copy the example environment file
cp .env.example .env
```
Edit the `.env` file with your Canvas credentials:
```env
CANVAS_API_TOKEN=your_canvas_api_token_here
CANVAS_BASE_URL=https://texastech.instructure.com
```
#### Getting Your Canvas API Token
1. Log in to Canvas at your institution's URL
2. Click your profile picture → **Settings**
3. Scroll to **Approved Integrations**
4. Click **+ New Access Token**
5. Enter a purpose (e.g., "Kiro CLI Integration")
6. Click **Generate Token**
7. **Copy the token immediately** — you won't see it again!
### 3. Configure Kiro CLI MCP Integration
Create or edit the MCP servers configuration file:
**File:** `~/.kiro/mcp_servers.json`
```json
{
"mcpServers": {
"canvas-lms": {
"command": "/ABSOLUTE/PATH/TO/canvas-lms-mcp/venv/bin/python",
"args": ["/ABSOLUTE/PATH/TO/canvas-lms-mcp/server.py"],
"env": {}
}
}
}
```
> **Important**: Replace `/ABSOLUTE/PATH/TO/canvas-lms-mcp` with the actual absolute path to your installation directory.
### 4. Test the Integration
Start Kiro CLI and test the Canvas integration:
```bash
kiro-cli chat
```
In the chat, you can now ask questions like:
- "What courses am I enrolled in on Canvas?"
- "Show me my upcoming assignments"
- "What announcements do I have?"
- "What's on my Canvas to-do list?"
## Available Canvas Tools
Once integrated, Kiro CLI has access to these Canvas tools:
| Tool | Description | Example Usage |
|------|-------------|---------------|
| `canvas_get_profile` | Get your Canvas profile | "Show me my Canvas profile" |
| `canvas_list_courses` | List enrolled courses | "What courses am I taking?" |
| `canvas_get_todo` | Get to-do items | "What's on my Canvas to-do list?" |
| `canvas_get_upcoming_events` | Get upcoming events | "What events do I have coming up?" |
| `canvas_get_assignments` | Get course assignments | "Show me assignments for course 58606" |
| `canvas_get_modules` | Get course modules | "What modules are in my CS course?" |
| `canvas_get_announcements` | Get announcements | "Any new announcements?" |
| `canvas_get_discussions` | Get discussion topics | "Show me discussion topics" |
| `canvas_get_grades` | Get grades | "What are my current grades?" |
## Example Conversations
### Getting Course Information
```
> What Canvas courses am I enrolled in?
I'll check your enrolled Canvas courses.
[Canvas API call results showing your active courses]
```
### Checking Assignments
```
> Show me upcoming assignments in my computer science courses
Let me check your Canvas assignments for CS courses.
[Canvas API call results showing assignments with due dates]
```
### Managing To-Do Items
```
> What's on my Canvas to-do list?
I'll get your Canvas to-do items.
[Canvas API call results showing pending tasks and due dates]
```
## Troubleshooting
### MCP Server Not Loading
1. **Check file paths**: Ensure all paths in `mcp_servers.json` are absolute
2. **Verify Python environment**: Make sure the virtual environment is properly created
3. **Check permissions**: Ensure the Python executable and server script are executable
### Authentication Errors (401)
1. **Verify API token**: Check that your Canvas API token is correct in `.env`
2. **Token expiration**: Canvas tokens may expire; generate a new one if needed
3. **Institution URL**: Ensure `CANVAS_BASE_URL` matches your institution's Canvas URL
### Canvas API Errors (403/404)
1. **Course access**: Some courses may be archived or you may not have access
2. **Permission levels**: Some endpoints require instructor privileges
3. **Course IDs**: Verify you're using correct course IDs from `canvas_list_courses`
### Environment Variable Conflicts
If you have `CANVAS_API_TOKEN` set as an environment variable, it may override the `.env` file:
```bash
# Check for conflicting environment variables
echo $CANVAS_API_TOKEN
# Unset if needed
unset CANVAS_API_TOKEN
```
## Advanced Configuration
### Custom Course Filtering
You can create a `test_hints.json` file to specify which courses to focus on:
```json
{
"valid_course_ids": [58606, 53482, 51243],
"test_assignment_id": null,
"test_module_id": null
}
```
### Logging and Debugging
To enable debug logging, modify the server startup in your MCP configuration:
```json
{
"mcpServers": {
"canvas-lms": {
"command": "/ABSOLUTE/PATH/TO/canvas-lms-mcp/venv/bin/python",
"args": ["/ABSOLUTE/PATH/TO/canvas-lms-mcp/server.py", "--debug"],
"env": {
"PYTHONPATH": "/ABSOLUTE/PATH/TO/canvas-lms-mcp"
}
}
}
}
```
## Security Considerations
- **API Token Security**: Your Canvas API token has full access to your account
- **Local Storage**: The `.env` file contains sensitive credentials
- **No Network Exposure**: The MCP server runs locally and doesn't expose network endpoints
- **Token Rotation**: Regularly rotate your Canvas API tokens for security
## Support
For issues specific to:
- **Canvas LMS MCP Server**: [GitHub Issues](https://github.com/sweeden-ttu/canvas-lms-mcp/issues)
- **Kiro CLI**: Amazon Web Services support channels
- **Canvas API**: Your institution's Canvas support
## Updates
To update the Canvas MCP Server:
```bash
cd canvas-lms-mcp
git pull origin main
source venv/bin/activate
pip install -e .
```
Restart Kiro CLI after updates to reload the MCP server.
