---
id: uplink
version: "0.10.0"
license: MIT License Copyright (c) 2018 P. Raj Kumar Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal… (full text in the JSON record)
license_treatment: permissive
maintenance: aging
---

# uplink — A Declarative HTTP Client for Python.

License: permissive · Maintenance: aging · Downloads: 190.7K/mo

## What it is and what it does

Uplink is a declarative HTTP client library that lets you define REST API clients as Python classes using decorators and type hints, inspired by Retrofit. Instead of writing ad-hoc requests calls, you annotate methods with @get, @post, etc., and specify path parameters, query strings, and request bodies through function arguments. It handles URL construction, parameter serialization, and response parsing automatically.

The library works with requests by default and supports async I/O through aiohttp and twisted. It can deserialize responses into custom Python objects, with optional integrations for marshmallow and pydantic schemas. You bring your own HTTP session for control over connection pooling, authentication, and other session-level concerns.

Use it for:

- Building a reusable Python client for a third-party REST API without writing repetitive request boilerplate.
- Creating internal API abstraction layers so application code calls clean Python methods instead of raw HTTP calls.
- Implementing async API clients with aiohttp or twisted when you need non-blocking I/O.
- Defining strongly-typed API contracts with decorators and type hints to catch errors early.
- Integrating with marshmallow or pydantic to automatically serialize and deserialize JSON responses.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Uplink turns HTTP APIs into declarative Python classes using decorators and type hints, letting you build reusable API clients with minimal boilerplate.

Yes, with conditions. Uplink is a solid choice for building structured REST clients in modern Python and has no known vulnerabilities. The low install friction and permissive MIT license make it safe to adopt. However, the aging maintenance status means you should verify that async support and optional integrations match your needs before committing to it for a new project.

## Install

pip install uplink
uv add uplink
poetry add uplink

## Installing uplink

Before you install: Low friction install with three lightweight runtime dependencies (requests, six, uritemplate). Maintenance status is aging—last release was 426 days ago—but the repository remains active with 1140 stars.

License in practice: MIT License permits unrestricted use, modification, and distribution with minimal obligations, making it safe for commercial and proprietary projects.

Quickstart:

pip install uplink

from uplink import Consumer, get, Path, Query

class GitHub(Consumer):
    @get("users/{user}/repos")
    def get_repos(self, user: Path, sort_by: Query("sort")):
        pass

github = GitHub(base_url="https://api.github.com/")
repos = github.get_repos(user="octocat", sort_by="created")

Requires Python 3.10 or later.

Verify before relying:

- Whether aiohttp and twisted async support are production-ready given the aging maintenance status
- Current state of optional marshmallow and pydantic integrations and their maintenance

## Package facts

- License: MIT License Copyright (c) 2018 P. Raj Kumar Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal… (full text in the JSON record) (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: aging
- Downloads: 190.7K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

declarative http client, api client generator, retrofit-style python, rest client decorator, structured api wrapper, http request abstraction, async api client, rest-client, decorator-based, async-ready

[View on SkillFed](https://skillfed.io/packages/uplink) · [View on PyPI](https://pypi.org/project/uplink/)