giskard-oss

An open-source evaluation and testing library for LLM agents.

🌍 OverseasFreeOpen source

Platforms: Self-hosted

Region: Overseas
Pricing: Free
Open source: Yes
GitHub Stars: ★ 5.4k
Source: GitHub
Added: 2026-06-06
Last verified: 2026-06-06
GitHub: github.com/Giskard-AI/giskard-oss

Overview

Giskard-oss is an open-source Python library designed for evaluating and testing LLM agents. It provides modular, lightweight, and dynamic tools to help developers ensure the quality and safety of their AI systems, particularly useful for testing multi-turn conversations or complex pipelines. Users can install the corresponding package and leverage the provided API to create custom test scenarios, enabling detection of model behavior under specific conditions.

Key features

▪Supports multi-turn conversation testing
▪Lightweight design reduces dependencies
▪Built-in multiple inspection functions
▪Asynchronous-first approach improves efficiency

Use cases

LLM model quality assessmentAI agent security vulnerability scanningGenerating synthetic datasetsRed team adversarial testing

Pros

＋Easy integration with existing projects
＋Highly flexible and extensible
＋Active community support

Limitations / notes

－Some advanced features are still under development

Who it's for

AI developersMachine learning engineersData scientistsSoftware testers

This overview was compiled by AI from public sources and may contain inaccuracies — please refer to the official site.

FAQ

How do I get started with Giskard-oss?

Install giskard via pip and follow the official documentation.

Which versions of Python does Giskard-oss support?

Requires Python 3.12 or higher.

Something wrong? Let us know on the About page and we'll fix it.