monitorit (awatch)

Self-hosted FastAPI monitoring with a built-in dashboard — traffic, errors, Apdex, request logs, consumers, uptime, and alerts.

No Grafana. No cloud account. No phone-home telemetry. Metrics stay on your machine in SQLite.

PyPI: monitorit · Import: from monitorit import awatch · Dashboard: /__awatch

from fastapi import FastAPI
from monitorit import awatch

app = FastAPI()
awatch.AWatch(app, env="dev")  # → http://127.0.0.1:8000/__awatch

A lightweight, privacy-first alternative to hosted API analytics (e.g. Apitally) and a simpler path than Prometheus + Grafana for single-app FastAPI services.

Why monitorit?

Install

Python 3.10+ recommended.

python3 -m venv .venv
source .venv/bin/activate          # Windows: .venv\Scripts\activate
pip install -U pip
pip install monitorit

From GitHub

pip install "git+https://github.com/Abdulwadood39/awatch.git"

Editable / contributing

git clone https://github.com/Abdulwadood39/awatch.git
cd awatch
pip install -e ".[dev]"

Verify

python -c "from monitorit import awatch; print(awatch.__version__, awatch.AWatch)"

Full details: docs/installation.md.

Quick start

1. Attach awatch to your FastAPI app (snippet above).
2. Run with uvicorn:

uvicorn your_module:app --reload

3. Open the FastAPI monitoring dashboard: http://127.0.0.1:8000/__awatch.

Try the demos in this repo

pip install -e ".[dev]"
uvicorn examples.basic_app:app --reload

curl http://127.0.0.1:8000/
curl -X POST http://127.0.0.1:8000/items \
  -H 'Content-Type: application/json' \
  -d '{"name":"gadget","price":9.5}'
curl http://127.0.0.1:8000/boom

Production auth

env="prod" requires an auth_token (or auth_dependency):

import os
from monitorit import awatch

awatch.AWatch(
    app,
    env="prod",
    auth_token=os.environ["AWATCH_TOKEN"],
    allow_ui_config=False,
)

Unlock the UI with ?token=... or the in-browser Unlock dialog. More: docs/usage.md.

Features

• Traffic / Errors / Performance — per-endpoint stats, Apdex, timelines
• Request inspector — headers, bodies, correlated logs, exceptions, cURL export
• Opt-in body/header logging with default secret masking
• Consumers — awatch.set_consumer() for individuals and groups
• Traffic labels — categories defined in code
• 422 heatmaps — which Pydantic fields fail validation
• Do not track — exclude sensitive paths (code or Settings)
• Uptime — heartbeat + synthetic checks + external ping
• Alerts — triggers in code → email / Slack / Discord / webhook
• Settings UI — SMTP, excludes, uptime, Apdex (lockable)
• Auth gate for production dashboards
• Quiet access logs — dashboard polls do not spam uvicorn (default on)

Documentation

Contributing

Contributions are welcome — bugs, docs, tests, dashboard UX, and features.

1. Open an issue for ideas or bugs
2. Fork, branch, and open a PR
3. Run pytest before submitting

Good first areas: documentation polish, more tests, Postgres storage, alert channels, dashboard UI.

Development

pip install -e ".[dev]"
pytest
uvicorn examples.basic_app:app --reload

Package layout: src/monitorit/awatch/ (core, capture, privacy, analytics, storage, health, triggers, api, dashboard).

Links

• GitHub: https://github.com/Abdulwadood39/awatch
• PyPI: https://pypi.org/project/monitorit/
• Issues: https://github.com/Abdulwadood39/awatch/issues
• Changelog: CHANGELOG.md

License

MIT © Abdulwadood