---
id: awsme
version: "1.0.0"
license: MIT license
license_treatment: permissive
maintenance: abandoned
---

# awsme — AWS CloudWatch client library to send metrics conveniently and efficiently

License: permissive · Maintenance: abandoned · Downloads: 97.9K/mo

## What it is and what it does

awsme is a CloudWatch metrics client library that wraps boto3 to simplify sending metrics to AWS CloudWatch. It supports both synchronous and asynchronous (threaded) metric sending, with optional buffering to batch metrics before transmission. By default, boto3 is an optional dependency, allowing the package to deploy with minimal overhead in AWS Lambda environments where the SDK is already available; outside Lambda, you install it via the boto3 extra.

The library lets you create a CloudWatch client with a namespace and default dimensions, then log individual metrics with optional per-metric dimensions, values, units, and storage resolution. It handles buffer flushing automatically at exit in standard applications, but provides an explicit flush method for Lambda functions where the process may be frozen before cleanup runs.

Use it for:

- Send application performance metrics to CloudWatch from a Python Lambda function without bundling boto3.
- Batch and asynchronously buffer metrics in a long-running application to reduce API call overhead.
- Attach consistent dimensions (e.g., version, environment) to all metrics from a service.
- Monitor custom application events (e.g., job completions, error counts) in CloudWatch dashboards.
- Test CloudWatch connectivity and AWS credentials from the command line using the included awsme-test tool.

## Worth the install?

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

Sends AWS CloudWatch metrics asynchronously and in batches from Python applications, with optional buffering and configurable dimensions.

No. The package is abandoned (last release 2019, no maintenance since early 2023) and depends on boto3 semantics that may have evolved. For new projects, use boto3 directly or a maintained CloudWatch wrapper. If you have legacy code already using awsme, it may still work, but expect no fixes for compatibility issues with modern Python or AWS SDK versions.

## Install

pip install awsme
uv add awsme
poetry add awsme

## Installing awsme

Before you install: Low install friction; depends only on click and attrs. Designed to deploy quickly in AWS Lambda by making boto3 optional. However, the package is abandoned—last release was 2019-03-07 and last commit 2023-02-10, so no maintenance or security updates are forthcoming.

License in practice: MIT license (permissive) allows commercial and private use with minimal restrictions, requiring only attribution.

Quickstart:

pip install awsme
# For use outside Lambda, add boto3:
# pip install awsme[boto3]

from awsme import create_cloud_watch
cloud_watch = create_cloud_watch(
    namespace='Application',
    dimensions={'version': '1.0.0'},
)
cloud_watch.log('metric', dimensions={'key': 'dim'}, value=123)

Requires AWS credentials configured via environment variables or configuration files (boto3 dependency must be installed if running outside AWS Lambda).

Verify before relying:

- Whether the package works correctly with modern versions of boto3, click, and attrs given its 2019 release date.
- Compatibility with Python versions beyond 3.6, which is the only version explicitly listed in classifiers.

## Package facts

- License: MIT license (permissive)
- Python support: unspecified
- Install friction: low
- Maintenance: abandoned
- Downloads: 97.9K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

cloudwatch metrics client, aws metrics logging, cloudwatch batch sender, asynchronous metrics buffering, aws monitoring library, cloudwatch python client, aws-cloudwatch, metrics-collection, abandoned

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