Giskard

Overview

Giskard is an open-source Python library for evaluating and testing LLM-based agents. It provides automated scanning for vulnerabilities like hallucinations, prompt injection, and bias, and integrates with existing CI/CD pipelines.

Best for

Best for
Python developers building LLM agents who need automated safety and quality testing.

Use cases

• Automated red-teaming of LLM agents for security flaws
• Regression testing LLM outputs across model versions
• Validating agent behavior against custom test suites

Notes

Giskard is an open-source Python library for evaluating and testing LLM-based agents. It provides automated scanning for vulnerabilities like hallucinations, prompt injection, and bias, and integrates with existing CI/CD pipelines.

5,414 stars on GitHub. Last updated 2026-05-29. Licensed Apache-2.0.

Use cases

• Automated red-teaming of LLM agents for security flaws
• Regression testing LLM outputs across model versions
• Validating agent behavior against custom test suites

Pros

• Comprehensive vulnerability scanning out of the box
• Active community with 5.4k GitHub stars
• Easy integration into Python testing workflows

Cons

• Limited to Python ecosystem only
• May require significant setup for complex agent architectures
• Documentation can be sparse for advanced use cases

Indexed from awesome-llm and enriched against its public facts.