Adwize

Open-source event monitoring with deterministic rules and webhook alerts. Ingest events from any source, define validation and threshold rules, and get notified instantly via webhooks when something goes wrong.

Features

• Event ingestion — Stream events via REST API with buffered bulk writes
• Three rule types — Threshold, Field Validation, and Volume anomaly detection
• Webhook alerts — Slack, Teams, or custom webhook notifications with retry
• CLI — Full management from the terminal with adwize commands
• MCP Server — AI-native interface for Cursor, Claude Desktop, and other MCP clients
• Self-hosted — Runs entirely on your infrastructure with Docker

Quick Start

One-liner install

curl -fsSL https://raw.githubusercontent.com/Adwize/adwize-oss/main/install.sh | bash

Then run the setup wizard:

cd ~/adwize
adwize setup

Manual setup

git clone https://github.com/Adwize/adwize-oss.git
cd adwize-oss
cp .env.example .env          # edit with your settings
docker compose up -d --build  # start API, worker, Postgres
uv sync                       # install CLI on your machine
source .venv/bin/activate     # activate to use `adwize` directly

The API will be available at http://localhost:8000 with interactive docs at /docs.

Note: docker compose runs the backend services. The CLI is a separate tool that runs on your machine and talks to the API. Install it with uv sync, then use adwize <command> or uv run adwize <command>.

Usage

Send events

adwize events send -s web -t page_view -d '{"url": "/pricing", "user_id": "u123"}'

Or via API:

curl -X POST http://localhost:8000/api/v1/events \
  -H "Content-Type: application/json" \
  -d '{
    "source": "web",
    "events": [{"type": "page_view", "data": {"url": "/pricing"}}]
  }'

Create rules

# Alert if more than 100 errors in 5 minutes
adwize rules create \
  -n "High error rate" \
  -t THRESHOLD \
  -c '{"event_type": "error", "field": "count", "operator": ">", "value": 100, "window_minutes": 5, "severity": "warning"}'

# Alert if purchase events are missing required fields
adwize rules create \
  -n "Purchase validation" \
  -t FIELD_VALIDATION \
  -c '{"event_type": "purchase", "required_fields": ["user_id", "amount", "currency"], "severity": "critical"}'

# Alert if page view volume drops 50% vs yesterday
adwize rules create \
  -n "Traffic drop" \
  -t VOLUME \
  -c '{"event_type": "page_view", "threshold_percent": -50, "current_window_minutes": 60, "severity": "critical"}'

Manage alerts

adwize alerts list                    # View all alerts
adwize alerts grouped                 # Grouped by rule
adwize alerts resolve <alert-id>      # Mark resolved
adwize alerts test-webhook --url https://hooks.slack.com/services/...

Live tail

adwize events tail                    # Stream all events
adwize events tail -s web -t error    # Filter by source and type

CLI Reference

adwize setup                    # Interactive setup wizard
adwize status                   # Health check + summary stats
adwize health                   # Quick health check
adwize up                       # Start services (docker compose up)
adwize down                     # Stop services (docker compose down)
adwize logs [-s api]            # View service logs
adwize mcp                      # Start MCP server (stdio mode)

MCP Server

Connect Adwize to Cursor, Claude Desktop, Windsurf, or any MCP-compatible AI assistant:

{
  "mcpServers": {
    "adwize": {
      "command": "adwize",
      "args": ["mcp"]
    }
  }
}

Or without the CLI on PATH:

{
  "mcpServers": {
    "adwize": {
      "command": "uv",
      "args": ["run", "python", "mcp_server/server.py"],
      "cwd": "/path/to/adwize-oss",
      "env": {
        "ADWIZE_API_URL": "http://localhost:8000"
      }
    }
  }
}

HLD

Configuration

The CLI stores config in ~/.adwize/config.json (created by adwize setup). Environment variables take priority.

Variable	Default	Description
DB_PASSWORD	required	PostgreSQL password
DB_NAME	adwize	Database name
DB_USER	postgres	Database user
API_KEY	(none)	Optional API key for auth
WEBHOOK_URL	(none)	Default webhook URL for alerts
ADWIZE_API_URL	http://localhost:8000	API base URL (CLI/MCP)
ADWIZE_API_KEY	(none)	API key for CLI/MCP

Development

uv sync                                         # Install dependencies
uv run uvicorn api.main:app --reload            # Run API locally
uv run python worker/rule_monitoring.py --once  # Run worker once
uv run python mcp_server/server.py              # Run MCP server

# Run MCP model inspector
ADWIZE_API_URL=http://localhost:8000 uv run mcp dev mcp_server/server.py

License

Apache License 2.0 — see LICENSE.