AI red teaming & LLM evaluation
Giskard is an open-source LLM agent testing library from Giskard AI. It is released under the Apache-2.0 license and is offered as: Library, Self-hosted, SaaS, On-prem.
Best for
Python teams that want a free library to scan an LLM agent or RAG app for prompt injection and harmful output, and to write pass or fail checks for it
Consider instead if
Giskard is an open-source Python library for testing and scanning LLM agents and RAG apps, made by Giskard AI. In version 3, its scan package generates adversarial test suites, runs them against your agent and prints a report grouped by threat type, with the judge's verdict and the full conversation for each failure. The docs list prompt injection, jailbreaks and harmful content among the attacks, and the scan includes multi-turn methods such as Crescendo and GOAT. The library is Apache-2.0 licensed and tests run on your machine. Giskard AI also lists an Enterprise plan with no published price. The README says version 2 is no longer actively maintained.
This summary is our own wording of the sourced facts on this page.
Each value links to the page that states it, with the date we checked it.
As documented by the vendor or project. The tag is our label from a controlled vocabulary.
Our editorial model, not a standard. The primary layer follows the category, except for tools whose main documented job is detection and response; other layers rest on a capability the sources document. How the model works.
| Layer | Role | How it applies | Based on |
|---|---|---|---|
| L4 Testing Testing and red teaming | Primary | Attack and evaluate models, apps and agents on purpose, before release and after every change, to find what the other layers miss. | Follows its category, AI red teaming & LLM evaluation |
The risk list is OWASP's (Top 10 for LLM Applications, 2025). The mapping is our analysis, and each row rests on a capability the sources document.
| Risk | How it applies | Based on |
|---|---|---|
| LLM01 Prompt Injection | Scans an agent for prompt injection and jailbreaks, including multi-turn attacks. | Scans for prompt injection, including a dedicated prompt injection generator and a GCG injection generator; Generates jailbreak attempts and requests for harmful content against an agent; Includes multi-turn attack generators Crescendo and GOAT |
| LLM02 Sensitive Information Disclosure | Homepage lists data disclosure among the security issues it looks for. | Looks for security issues such as data disclosure before production |
| LLM09 Misinformation | Targets hallucination and checks answers against a knowledge base. | Targets hallucination and checks answers against a knowledge base to catch invented policies |
More in AI red teaming & LLM evaluation, ranked within the category by Coverage Score.
NVIDIA
Probes language models and LLM apps for prompt injection, jailbreaks, data leakage and more.
Best for: Teams that want a free command-line scanner they can point at any model or LLM endpoint before release
Reviewed
OpenAI (Promptfoo)
Runs evals and red team scans against LLM apps from a CLI or CI pipeline and writes a vulnerability report.
Best for: Developer teams that want one open-source CLI for both prompt evals and red teaming, run locally and in CI
Reviewed
Praetorian
Single-binary Go scanner that runs jailbreak, prompt injection and other attack probes against LLMs.
Best for: Security teams that want a free single-binary scanner to probe models and LLM endpoints with jailbreaks, prompt injection and multi-turn attacks
Reviewed
Yes. The Giskard library is released under the Apache-2.0 license, and the code is on GitHub at Giskard-AI/giskard-oss. The pricing page lists a free plan for the open-source library.
Version 3, the rewrite for testing agents. The README says version 2 remains available but is no longer actively maintained, and that v3 has no equivalent for the v2 tabular model scan.
The pricing page lists a free plan for the open-source library at 0 USD and an Enterprise plan with no price shown and a contact link.
Giskard is released under the Apache-2.0 license.
Source Source, checked October 21, 2026: github.com (opens in a new tab) Checked
Giskard is offered as: Library, Self-hosted, SaaS, On-prem.
Source Source, checked October 21, 2026: giskard.ai (opens in a new tab) Checked
Pricing model for Giskard: Freemium: Free plan: open-source library, 0 USD. Enterprise plan: no price listed, with a contact link..
Source Source, checked October 21, 2026: giskard.ai (opens in a new tab) Checked
Latest release of Giskard: October 20, 2026.
Source Source, checked October 21, 2026: api.github.com (opens in a new tab) Checked
We have no commercial relationship with Giskard AI.