AI red teaming & LLM evaluation
Agentic Security is an open-source LLM vulnerability scanner from msoedov (open-source maintainer). It is released under the Apache-2.0 license and is offered as: CLI, Self-hosted.
Best for
Developers who want a free scanner they can run locally or in CI to replay public jailbreak datasets against an LLM API or agent endpoint
Agentic Security is an open-source (Apache-2.0) vulnerability scanner for LLM endpoints and agent workflows. You install it with pip and run it as a command-line tool, a local web UI or a CI step. You describe the target as an HTTP request with a prompt placeholder, and the scanner replays attack datasets from Hugging Face, public CSV files or local files against it, with text, image and audio inputs. It can mutate prompts with encodings such as rot13 and base64, and it has a configuration option for multi-step jailbreaks. The docs site says it is a test tool, so a clean scan does not prove a target is safe. The repository is owned by an individual maintainer, not a company.
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 | Replays jailbreak datasets and multi-step jailbreak attacks against the target. | Replays jailbreak prompt datasets from Hugging Face and public CSV files against the target; Offers multi-step jailbreak attacks, switched on by a configuration option |
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
Giskard AI
Generates adversarial test suites for LLM agents and RAG apps, runs them and reports failures by threat type.
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
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
Yes. It is licensed Apache-2.0 and the code is on GitHub at msoedov/agentic_security. You install it with pip.
Any LLM HTTP API that you describe as a request template with a prompt placeholder. The README shows examples for text, image and audio inputs.
No. The docs site calls it a test tool and says a clean scan does not prove a target is safe.
Agentic Security is released under the Apache-2.0 license.
Source Source, checked October 21, 2026: github.com (opens in a new tab) Checked
Agentic Security is offered as: CLI, Self-hosted.
Source Source, checked October 21, 2026: github.com (opens in a new tab) Checked
Pricing model for Agentic Security: Free.
Source Source, checked October 21, 2026: github.com (opens in a new tab) Checked
Latest release of Agentic Security: June 3, 2026.
Source Source, checked October 21, 2026: api.github.com (opens in a new tab) Checked
We have no commercial relationship with msoedov (open-source maintainer).