Compare vendors / Giskard (open source)
Giskard (open source)
Open-source evals and red teaming library for agents
Agents and toolsOpen source
Giskard’s open-source library covers evals, red teaming and test generation for agentic systems. Version 3, a ground-up rewrite (3.0.1 on PyPI, Oct 2026), splits into giskard-checks and giskard-scan, whose vulnerability scan generates adversarial inputs across OWASP LLM Top 10 categories. It has about 5,900 GitHub stars.
Best for: Teams that want open-source agent tests with a paid upgrade path.
Open source (Apache-2.0), maintained by Giskard
- ✓ Maps to OWASP
- – Maps to MITRE ATLAS
- – Maps to the EU AI Act
- – SOC 2 or ISO 27001
1 Legal nameGiskard (open-source library)
2 HeadquartersNot disclosed
3 Founded2022
4 TeamNot disclosed
5 What they testLLM apps, Agents and tool use, RAG, Prompt injection and jailbreaks, Data leakage, Bias and harms
6 FrameworksOWASP LLM Top 10
7 CertificationsNot disclosed
8 Published pricesFree (Apache-2.0). Giskard Hub is the paid enterprise product.
9 Customers and partnersNot disclosed
Worth knowing before you sign
- v2, including the original LLM scan and RAGET, is no longer maintained; v3 needs Python 3.12+.
- The 50+ probes figure applies to the paid Hub, not the open-source scan.
Go to Giskard (open source)'s site
Sources
Work at Giskard (open source)? Send corrections with a source to editors@bestairedteamingcompanies.com. We fix verified errors within five business days.
Compare with
- Mindgard AI red teaming firm: human-led tests plus an automated platform
- Gray Swan AI AI security lab: agent red teaming, lab evals and the Arena
- HiddenLayer AI security platform with automated red teaming and model scanning
- Trail of Bits Security research consultancy with an AI/ML assurance practice
- Pillar Security Agent security platform with agentic red teaming