README.md

Forest is the back office for operations teams in regulated companies. Ops teams review, decide and act on your data from one interface, or let AI agents do it, within the roles, permissions and audit logs you define.

This repository is the Python agent for Django and Flask: the part of Forest that runs inside your application, connects to your databases, and exposes them to the Forest interface. Your business data never goes through Forest.

Note

Forest v2. This is the current Python SDK. Using django-forestadmin? You are on v1: see the migration guide.

Quick start

Django

pip install forestadmin-agent-django

The agent plugs in through your settings, not a constructor:

# settings.py
import os

INSTALLED_APPS = [
    # ...
    "forestadmin.django_agent",
]

FOREST_AUTH_SECRET = os.environ.get("FOREST_AUTH_SECRET")
FOREST_ENV_SECRET = os.environ.get("FOREST_ENV_SECRET")

# urls.py
from django.urls import include, path

urlpatterns = [
    path("", include("forestadmin.django_agent.urls")),
]

Your Django models are exposed automatically. To add computed fields, actions or segments, point FOREST_CUSTOMIZE_FUNCTION to a function that receives the agent (e.g. "myapp.forest_admin.customize_forest"). The module must exist: if it cannot be imported, the agent does not start.

Flask + SQLAlchemy

pip install forestadmin-agent-flask forestadmin-datasource-sqlalchemy flask-sqlalchemy

import os

from flask import Flask
from flask_sqlalchemy import SQLAlchemy
from forestadmin.datasource_sqlalchemy.datasource import SqlAlchemyDatasource
from forestadmin.flask_agent.agent import create_agent

db = SQLAlchemy()  # your models are declared on this instance

def create_app():
    app = Flask(__name__)
    app.config["SQLALCHEMY_DATABASE_URI"] = os.environ["DATABASE_URL"]
    # the agent reads its settings from app.config, not from the environment
    app.config["FOREST_ENV_SECRET"] = os.environ["FOREST_ENV_SECRET"]
    app.config["FOREST_AUTH_SECRET"] = os.environ["FOREST_AUTH_SECRET"]

    agent = create_agent(app)
    with app.app_context():
        db.init_app(app)
        agent.add_datasource(SqlAlchemyDatasource(db))
    agent.start()
    return app

Full guide: docs.forest.app/get-started/quickstart

Requirements

Python	>=3.8,<3.14 (CI tests 3.8 to 3.13)
Django	>=3.2,<5.2
Flask	>=2.0.0 (async extra)
SQLAlchemy	>=1.4.0

What the agent gives you

Data	Your Django models or SQLAlchemy tables, introspected and exposed to Forest.
Customization	Computed fields, actions, segments and hooks in Python, from your customize function.
Permissions and audit	Roles, scopes and activity logs defined in Forest, enforced on every request.

How it works

The agent runs on your infrastructure. The Forest interface calls your agent, and your agent queries your sources: Forest does not store your business data.

Documentation

Get started	https://docs.forest.app/get-started/quickstart
Connect a data source	https://docs.forest.app/get-started/connect/data-sources/overview
Migrating from v1	https://docs.forest.app/guides/migration/from-v1/overview

Packages in this monorepo

PyPI package	Role
forestadmin-agent-django	Agent for Django
forestadmin-agent-flask	Agent for Flask
forestadmin-agent-toolkit	Agent core (transport, auth, permissions)
forestadmin-datasource-toolkit	Data source contracts, queries, types
forestadmin-datasource-django	Django ORM data source
forestadmin-datasource-sqlalchemy	SQLAlchemy data source

All packages install into the shared forestadmin.* namespace. Complete demo applications live in src/_example/.

Contributing

Bug reports and pull requests are welcome. See CONTRIBUTING.md for setup, tests and commit conventions.

./launch_tests_ci_like.sh   # runs the CI locally

License

Apache 2.0. See LICENSE.