CheckOps
Agentic Security
Agentic Security is an Apache-2.0 open-source vulnerability scanner for LLMs and agent workflows that probes jailbreaks, fuzzing and multimodal attacks across text, images and audio.
- Open source
- Free
- Python
- Fuzzing
- Ai Red Teaming
- Ai Security
- LLM Testing
- AI-enabled testing
- Security testing
- Release & CI gates
Latest public signals
Not refreshed yet. Refresh reads the public site and, when the entry has a GitHub repository, the latest release.
Introduction
Agentic Security is an open-source scanner for agent workflows and large language models. It protects AI systems against jailbreaks, fuzzing and multimodal attacks.
It probes text, image and audio inputs, simulates iterative multi-step attack sequences, stress-tests models with randomised inputs, connects to LLM APIs for high-volume attack scenarios, and uses reinforcement learning to craft adaptive probes that evolve with a model’s defences.
The source code is at msoedov/agentic_security on GitHub under the Apache-2.0 licence, with about 2,000 stars; the most recent commit was on 11 November 2026.
Features
- Multimodal probing: Text, images and audio.
- Multi-step attacks: Iterative jailbreak sequences.
- Fuzzing: Randomised inputs.
- API integration: Test any LLM endpoint.
- RL-driven probes: Adaptive attacks.
Pros
- Free to use: Open source under the Apache-2.0 licence, so the core tool costs nothing to use.
- Actively maintained: The repository had commits as recent as 11 November 2026.
Cons
- Smaller project: Fewer maintainers than garak.
- Python setup: Requires local environment.
Usage and Links (Example)
pip install agentic_security
agentic_security --help
agentic_security init # create a CI config
agentic_security ci # run scans in CI
Running agentic_security starts a local UI (set the port with --port) where you point it at your LLM endpoint.
Resources:
Pricing
Prices were not listed on the website when checked on 11 November 2026; contact the vendor for plans.
Recommended for?
- AI engineers adding jailbreak scans to CI.
Conclusion
Agentic Security is a capable open-source LLM attack scanner.
Run it alongside garak for broader probe coverage.
On CheckOps, Agentic Security is filed under the Security testing and Release & CI gates stages. Related tools listed here include garak, PyRIT, DeepTeam, Promptfoo, Giskard and Lakera. When comparing them, weigh the licence, the languages and platforms each supports, whether it is self-hosted or managed, and how cleanly it slots into your existing CI pipeline. Details on this page were checked against the official website, documentation and repository on 11 November 2026.
Alternatives
Other tools on CheckOps that cover similar ground. Open a page, or compare one with Agentic Security side by side.
Ggarakgarak is NVIDIA's Apache-2.0 LLM vulnerability scanner that probes models and dialog systems for hallucination, data leakage, prompt injection, toxicity, jailbreaks and more.Compare with Agentic Security
PPyRITPyRIT is Microsoft's MIT-licensed Python Risk Identification Tool for generative AI: automated scans, adaptive multi-turn attacks and human-led red teaming in one toolkit.Compare with Agentic Security
DDeepTeamDeepTeam is an Apache-2.0 open-source red-teaming framework from Confident AI that simulates jailbreaks, prompt injection and multi-turn attacks to find vulnerabilities in LLM apps and agents.Compare with Agentic Security
PPromptfooPromptfoo is an open-source CLI for LLM evals and red teaming. The community tool is free; enterprise features are quoted.Compare with Agentic Security
GGiskardGiskard offers an Apache-2.0 open-source library for evals, red teaming and test generation for AI agents, plus a commercial Hub for agent red-teaming, evaluation and guardrails.Compare with Agentic Security
LLakeraLakera, now part of Check Point, offers AI security products: a Guard API (documented as Check Point AI Guardrails) for runtime screening of prompt injection and data leaks, and AI red teaming.Compare with Agentic Security