What is the benefit of using vcr-http-recording?

It makes tests faster and more reliable by replacing real network calls with recorded responses, preventing test flakiness and significantly reducing API usage costs.

How does it handle sensitive information like API keys?

The skill provides built-in configurations to automatically filter and redact sensitive headers, query parameters, and body content before they are saved to your test cassettes.

Does this skill support asynchronous Python code?

Yes, it includes specialized patterns and configurations for recording interactions from popular async clients such as httpx and aiohttp.

Can I use this for testing LLM APIs like OpenAI or Anthropic?

Absolutely. It features custom matchers designed to ignore dynamic fields like timestamps or request IDs in LLM requests, ensuring stable and repeatable test results.

Is it safe to use in CI/CD pipelines?

Yes, it recommends using a 'none' record mode in CI environments to ensure tests only use existing cassettes and never attempt to make unauthorized live network calls.

VCR HTTP Recording for Python

Name: VCR HTTP Recording for Python
Author: yonatangross

byyonatangross

•

Security & Testing

Automates the recording and replaying of HTTP interactions to create deterministic, fast, and network-independent Python tests.

This skill enables Claude to implement VCR.py patterns within Python projects, allowing developers to capture real API responses and store them as 'cassettes.' It provides standardized configurations for pytest, support for asynchronous requests with httpx or aiohttp, and advanced techniques for redacting sensitive credentials like API keys. By replaying recorded interactions, it ensures tests remain consistent, avoids hitting external rate limits, and allows for offline development while specifically addressing the complexities of testing LLM-based applications through custom request matching.

Key Features

0169 GitHub stars

02Full support for async Python clients like httpx and aiohttp

03Configurable recording modes optimized for CI/CD environments

04Deterministic HTTP mocking using VCR.py cassettes

05Custom request matching logic for dynamic payloads and LLM prompts

06Automated redaction of sensitive API keys and authorization headers

Use Cases

01Testing LLM-integrated applications without incurring repetitive API costs

02Creating offline-capable integration tests for external REST APIs

03Standardizing mock data across a development team to ensure test consistency

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add yonatangross/orchestkit vcr-http-recording

For use in Claude.ai and ChatGPT

Download Skill