Projects represent individual applications or environments within your organization. Each project has its own API keys, logs, and settings.
Creating a Project
- Navigate to your organization
- Click New Project
- Enter a project name (e.g., “Production”, “Chat App”, “Staging”)
- Click Create
Project Structure
Each project contains:
- API Keys - Muxx keys for authentication
- Provider Keys - Your LLM provider API keys
- Logs - Request history for this project
- Analytics - Usage metrics and costs
- Settings - Configuration and limits
Project Settings
General
- Name - Project display name
- Description - Optional description
- Environment - Production, Staging, Development (for labeling)
Provider Keys
Add your LLM provider API keys:
- Go to Settings → Provider Keys
- Click Add Key
- Select provider (OpenAI, Anthropic, Google)
- Paste your API key
- Click Save
Provider keys are encrypted at rest and never logged in request data.
Gateway Settings
Configure gateway behavior:
- Caching - Enable/disable, set TTL
- Rate Limiting - Requests per minute, tokens per minute
- Spend Caps - Daily/monthly limits
Alerts
Set up notifications for:
- High error rates
- Approaching rate limits
- Spend thresholds
Switching Projects
Use the project dropdown in the header to switch between projects within an organization.
Deleting a Project
Deleting a project permanently removes all logs, analytics, and settings. This cannot be undone.
- Go to Settings → Danger Zone
- Click Delete Project
- Type the project name to confirm
- Click Delete
Project Patterns
By Environment
Organization: Acme Corp
|-- Production
|-- Staging
|-- Development
By Application
Organization: Acme Corp
|-- Chat App
|-- Document Processor
|-- Customer Support Bot
Combined
Organization: Acme Corp
|-- Chat App - Production
|-- Chat App - Staging
|-- Support Bot - Production
|-- Support Bot - Staging
API Key Per Project
Each project has its own Muxx API key. This means:
- Logs are automatically separated by project
- Rate limits apply per project
- Different provider keys per project (e.g., different OpenAI orgs)