Code-and-Sorts/PromptDrifter

🧭 PromptDrifter – one‑command CI guardrail that catches prompt drift and fails the build when your LLM answers change.

38
/ 100
Emerging

PromptDrifter helps developers prevent unexpected issues in applications that use Large Language Models (LLMs). It takes your defined prompts and expected LLM responses, then automatically checks if the LLM's answers have changed. If the responses deviate from what you expect, it flags the issue early in your development process, before it impacts users. This tool is for developers building and maintaining applications powered by LLMs.

Available on PyPI.

Use this if you need to ensure that your LLM-powered application consistently produces the expected outputs, even as underlying LLM models or versions change.

Not ideal if you are not integrating LLMs into an application and do not have a continuous integration (CI) pipeline.

LLM application development prompt engineering software testing continuous integration AI product reliability
Maintenance 10 / 25
Adoption 4 / 25
Maturity 24 / 25
Community 0 / 25

How are scores calculated?

Stars

7

Forks

Language

Python

License

MIT

Last pushed

Mar 05, 2026

Commits (30d)

0

Dependencies

11

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/prompt-engineering/Code-and-Sorts/PromptDrifter"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.