For AI agents: A markdown version of this page is available at https://docs.datadoghq.com/security/application_security/setup/compatibility/python.md. A documentation index is available at /llms.txt.

Python Compatibility Requirements

This product is not supported for your selected Datadog site. ().

App and API Protection capabilities

The following App and API Protection capabilities are supported in the Python library, for the specified tracer version:

App and API Protection capabilityMinimum Python tracer version
HTTP Monitoring1.16.0
gRPC Monitoringnot supported
GraphQL Monitoringnot supported
Exploit Prevention2.20.0
API Security4.6.0
Account Takeover Protection2.11.0
Runtime Activation1.6.0
Runtime Software Composition Analysis (SCA)1.5.0
Runtime Code Analysis (IAST)3.15.0

The minimum tracer version to get all supported App and API Protection capabilities for Python is 4.6.0. The following capabilities are not yet supported: gRPC Monitoring, GraphQL Monitoring. Note: AAP requires enabling Remote Configuration, which is included in the listed minimum tracer version.

Supported deployment types

Deployment typeSupport
Docker
Kubernetes
Amazon ECS
AWS Fargate
AWS Lambda

Language and framework compatibility

Supported Python versions

The Python App and API Protection Client library follows a versioning policy that specifies the support level for the different versions of the library and Python runtime.

Two release branches are supported:

ReleaseSupport levelMinimum Datadog Agent
>=4.0,<5General Availability7.28
>=3.0,<4Maintenence (until May 31, 2026)7.28
>=2.0,<3End of Life7.28
<2End of Life

And the library supports the following runtimes:

OSCPURuntimeRuntime versionSupported ddtrace versions
Linuxx86-64, AArch64CPython3.9+>=4, <5
MacOSIntel, Apple SiliconCPython3.9+>=4, <5
Windows64bit, 32bitCPython3.9+>=4, <5
Linuxx86-64, i686, AArch64CPython3.8+>=3, <4
MacOSIntel, Apple SiliconCPython3.8+>=3, <4
Windows64bit, 32bitCPython3.8+>=3, <4
Linuxx86-64, i686, AArch64CPython3.7-3.13>=2,<3
MacOSIntel, Apple SiliconCPython3.7-3.13>=2,<3
Windows64bit, 32bitCPython3.7-3.13>=2,<3
Linuxx86-64, i686, AArch64CPython2.7, 3.5-3.11<2
MacOSIntel, Apple SiliconCPython2.7, 3.5-3.11<2
Windows64bit, 32bitCPython2.7, 3.5-3.11<2

Integrations

The Python tracer includes support for the following frameworks, data stores and libraries.

If you don't see your library of choice listed, fill out this short form to send details.

Web framework compatibility

  • Attacker source HTTP request details
  • Tags for the HTTP request (status code, method, etc)
  • Distributed Tracing to see attack flows through your applications
Tornado integration requires setting the DD_TRACE_TORNADO_ENABLED environment variable to true.
FrameworkHTTP MonitoringHTTP Blocking
Django
Flask
FastAPI
Tornado

User Authentication Frameworks compatibility

Integrations to User Authentication Frameworks provide:

  • User login events, including the user IDs
  • Account Takeover detection monitoring for user login events
FrameworkAutomatic User Event Tracking
Django

Networking framework compatibility

FrameworkServer-side Request Forgery (SSRF)
Requests
urllib3
httpx
urllib
http.client
webbrowser

Data store compatibility

Datastore tracing provides:

  • SQL attack detection
  • Query info (for example, a sanitized query string)
  • Error and stacktrace capturing

Note: App and API Protection also analyzes untrusted HTTP request input, so some protections can still apply even when a database integration is not listed here.

FrameworkSQL Injection (SQLi)
psycopg
mysql-connector-python
mysqlclient
PyMySQL
sqlite3
MariaDB
pyodbc
Snowflake Connector

Filesystem compatibility

FrameworkLocal File Inclusion (LFI)
pathlib
builtins

Subprocess compatibility

FrameworkCommand InjectionShell Injection
subprocess
os