AI agent security guard that stops threats instantly.

Open

May 20, 2026

2026 Rank: #675

Adrian

United Kingdom AI security

5,136

5.0(2)

Use tool Copy 🔗

5,136

5.0(2)

Inputs:

Outputs:

Agents MCP (Model Context Protocol)No Signup Open Source Local

AI agent security guard that stops threats instantly.

Local No Sign-Up AI Security Runtime Monitoring AI Behavior Analysis AI AgentFree

Overview

Overview Releases Alternatives Pricing Pros & Cons Prompts Reviews Q&A

Featured alternatives

Overview Discussion

4,649 github.com

Overview

Socials:

Adrian is an open-source runtime security monitoring and control tool that has been designed specifically for Artificial Intelligence (AI) agents. Coming from the secureagentics repository on GitHub, Adrian is ingeniously designed to detect and control malicious or misaligned behavior of AI agents during their operation.

Key features include the ability to monitor an agent's activity logs, tool calls, outputs, actions, and reasoning traces. By scrutinizing these traces and logs, the tool can identify deviant behaviors and react in real-time, even before the agent acts.

One distinguishing aspect is that it allows for optional in-flight intervention for behaviour that deviates from the expected or acceptable range.For developers and AI practitioners, Adrian is easily accessible and straightforward to utilize.

It is available through a Python Software Development Kit (SDK), which can be installed and integrated with just two lines of code dedicated to LangChain agents.

This real-time monitoring system allows developers and AI practitioners to ensure their AI tools and models are acting within their defined remit, hence reducing the risk of unexpected or unwanted behaviors.

Adrian offers two modes of operation: audit and block mode. The audit mode is designed for general monitoring of agent behavior, while the block mode provides the added security measure of intervening and stopping any suspicious activity.

The customization extends towards the alert systems, which allows users to set up specific alerting channels and define accepted behaviours versus known risks.In sum, Adrian provides a comprehensive security and control system for the runtime operations of AI agents.

It ensures that these agents act according to the behaviors that are programmed in them, thereby reducing the risk of malicious activity or policy drift.

Supported features

Releases

AdrianInitial

Get notified when a new version of Adrian is released

Notify me

Initial release

May 20, 2026

Initial release of Adrian.

Author

Max Corbridge

@max-corbridge

Ethical hacker turned founder. Now CEO & Cofounder @ Secure Agentics

secureagentics.ai

🇬🇧 United Kingdom

Stats

1 tool

Beginner

Joined: May 2026

Pricing

Pricing model

Free

Paid options from

Free

Keeping you safe

Good to know

Terms & Conditions

Use tool

Save

🔗 Copy link

🗳️ Vote Best AI Tool

Featured

AI security Adrian

United Kingdom AI security

5,136

5.0(2)

Overview Releases Alternatives Pricing Pros & Cons Prompts Reviews Q&A

Use tool

Save

Top alternatives

AI Security Gateway

AI Firewall & LLM Proxy — PII Redaction, Smart Routing, Budget Enforcement

7,118 opensourceaihub.ai

ai arena

🙏 3 karma

Apr 14, 2026

@AI Security Gateway

AI firewall for PII and smart routing looks promising.

3 Reply Share Edit Delete Report

Share

Released 1mo ago
Free + from $29/mo

7,646
7
4.6
iDox.ai Guardrail

Stop sensitive data leaks to AI tools.

5,997 www.idox.ai

Share

🇺🇸 United States
Released 22d ago
Free + from $10/mo

6,140
4
Exogram

The execution trust layer for autonomous AI

4,859 exogram.ai

Share

Released 28d ago
Free + from $99.00/mo

5,217
4
5.0
Liminal

Secure generative AI without compromising data

Share

Released 1y ago
No pricing

548
1
TrojAI

Secure your AI models from risks and attacks.

Share

Released 2y ago
No pricing

396
1
HiddenLayer

The most comprehensive AI security platform.

180 www.hiddenlayer.com

Share

Released 3y ago
No pricing

223

Reviews

5.0

Average from 2 ratings.

★ ★ ★ ★ ★ 2

★ ★ ★ ★ 0

★ ★ ★ 0

★ ★ 0

★ 0

Your rating

★ ★ ★ ★ ★

Attach prompt

Attach result

Post

How would you rate Adrian?

Help other people by letting them know if this AI was useful.

Prompts & Results

Title:

Description:

Prompt type:*

Prompt:*

Output type:*

Output:*

Add your own prompts and outputs to help others understand how to use this AI.

Pros and Cons

Pros

Real time security monitoring

Direct control over agents

Analyzes agent activity logs

Analyzes reasoning traces

Can intervene during events

Python SDK integration

Two operation modes: audit/block

Alerting channels feature

Defines acceptable behaviours

Highlights known risks

Self-hosting compatibility

Open-source tool

Detects policy drift

Prevents threat attempts

Behavior analysis of agents

In-flight behaviour intervention

Customizable alert system

Audit mode for monitoring

Block mode for intervention

Generates security audit

Hosted on GitHub

Real-time threat response

Agent activity log monitoring

Accessible and easy use

Python SDK: simple integration

Low risk of unwanted behaviours

Designed for range of behaviours

Flexible alert customization

Dashboard for real-time updates

Can install with two lines

Monitors tool calls/outputs/actions

Real-time classified events

Offline, data sovereign deployments

Entire stack runs on single host

Reasoning analysis boosts detection accuracy

Free, open-source tool

Can run on user's infrastructure

View 32 more pros

Cons

Python SDK restricted

Customization complexities

Limited alert channels

Dependent on LangChain agents

Self-hosting pains

Requires frequent interventions

Offline deployments demanding

Single host limitations

View 3 more cons

Q&A

What is Adrian and what does it do?

Adrian is a runtime security monitoring and control engine designed for AI agents. It is capable of identifying and handling security matters such as the misuse of tools, prompt injections, and policy drifts in real time, before the AI agent conducts the action. It does this by scrutinizing both agent activity logs, encompassing tool calls, actions, and outputs, as well as reasoning traces to spot any potential harmful or misaligned behavior.

How can Adrian detect harmful or misaligned behavior in AI agents?

Adrian employs a detailed analysis of both agent activity logs and reasoning traces to detect harmful or misaligned behavior. The agent activity logs include tool calls, actions, and outputs carried out by the AI. On the other hand, reasoning traces allow Adrian to understand the logic or reasons behind these actions. By combing the two, it can identify any behaviour that may pose a security risk.

What are the features of Adrian?

Adrian offers a suite of features geared towards ensuring AI agent security. These include real-time analysis of agent activities and reasoning traces, malicious tool use detection, real-time understanding and intervention of policy drift, prompt injection, and more. It also provides alerting channels, the ability to delineate acceptable behaviours and known risks, and compatibility for self-hosting on users' own infrastructure. Furthermore, it facilitates easy integration into LangChain agents with its Python SDK.

Can Adrian interrupt potentially harmful AI activities?

Yes, Adrian is designed not only to monitor AI activities but also to intervene them when necessary. This is mainly done in its block mode operation. When a potential harmful activity is identified, Adrian can interrupt the suspect activity as part of its control measures.

How to integrate Adrian with LangChain agents using Python SDK?

Integration of Adrian into LangChain agents is made seamless with its Python Software Development Kit (SDK). You first need to install the SDK using pip, the Python package installer. Then a LangChain provider for the agent's model needs to be installed. Finally, an 'adrian.init' and 'adrian.shutdown' bracket is wrapped around the standard LangChain.

What is the difference between Adrian's audit mode and block mode?

In Adrian, the audit mode is responsible for identifying and classifying security events. It observes AI agent operations and raises an alert when it spots activities that may pose a risk based on set guideline. On the other hand, block mode goes a step further. When suspicious activities are identified, block mode interrupts these actions rather than just generating an alert.

+ Show 34 more

Can Adrian be self-hosted, and if so, how?

Yes, Adrian can be hosted on a user's own infrastructure. This ensures data privacy and sovereignty. Adrian supports entirely offline, data sovereign deployments using just a handful of Docker commands. The Go backend, the Next.js dashboard, the Python SDK, and a Llama.cpp container that serves a local Gemma model can all be run on a single host.

Does Adrian offer real-time analysis and security event classification?

Absolutely, Adrian offers real-time analysis and security event classification. By continuously monitoring agent activity logs and reasoning traces, it can immediately detect any aberrations or behaviors that violate set guidelines. Once identified, these security events are promptly classified depending on their nature and severity for appropriate handling.

How can I identify and delineate acceptable behaviors in Adrian?

Adrian provides its users the ability to establish acceptable behaviours for their AI agents. This is done during configuration, where one sets the remit of the agent, distinguishes between the behaviors that are permissible versus the known risks. With these boundaries in place, Adrian can then detect any deviations and take corrective actions when an AI agent engages in unacceptable behaviours.

Is Adrian capable of detecting policy drift in real time?

Yes, Adrian is designed to detect policy drift in real time. Its continuous monitoring and control system is adept at catching changes that indicate a drift from pre-established policies. This allows it to intervene and correct course as soon as it identifies the deviation.

What are the alerting channels available in Adrian?

Adrian supports alerting channels, making sure notifications about identified security events reach you. The specific types of alerting channels are not explicitly stated on their website presently. However, they could potentially include email alerts, SMS notifications, system messages, etc.

How can Adrian prevent the misuse of AI tools?

Adrian is geared towards preventing misuse of AI tools. It does this by continuously monitoring and analyzing two main things: agent activity logs and reasoning traces. If it detects suspicious activities, like malicious tool use, prompt injection, or policy drift, it can intervene in real time and block these activities before they're executed.

What does Adrian's block mode actually do?

Block mode is an operational mode in Adrian that not only identifies the malicious activities of an AI agent but also interrupts them. When Adrian operates in block mode, it stops the identified suspect activities, providing an extra layer of security by taking proactive actions against potential risks.

Can Adrian analyze agent activity logs and reasoning traces?

Yes, Adrian is capable of analyzing both agent activity logs and reasoning traces. While activity logs provide a record of the actions, output, and tool calls made by an AI agent, reasoning traces help in understanding the context, logic, or rationale behind these actions. This dual analysis enhances its ability to detect irrational behavior.

How can Adrian contribute to malware prevention?

Adrian contributes to malware prevention by spotting and acting on malicious or misaligned behaviours before they're executed. This preemptive approach identifies harmful actions, like misuse of tools, in real time and interrupts the agent's activities if they pose threat; thus, mitigating the risk of malware-related issues.

Can Adrian help in managing known risks and risk prevention?

Adrian plays a proactive role in risk management. It delineates acceptable behaviours and known risks, helping the user to clearly define what the AI agent is and isn't permitted to do. If the agent attempts to deviate from established guidelines or falls into known risk patterns, Adrian detects this and can intervene thereby preventing the risk from escalating.

How can I integrate Adrian into my own infrastructure?

Integrating Adrian into your own infrastructure is a straightforward process facilitated by its compatibility for self-hosting. Users need to use Docker commands to run the Go backend, the Next.js dashboard, the Python SDK, and a Llama.cpp container that serves a local Gemma model on a single host machine. This way, Adrian can be made to operate on your own infrastructure, ensuring data sovereignty and privacy.

What are the potential use cases for Adrian?

Adrian provides a flexible solution for numerous use cases that require real-time AI security monitoring and control. These could range from use cases where there need to be monitoring and control of AI agents in e-commerce platforms, customer service chatbots, automated data analysis systems, healthcare AI applications, AI-driven utility management systems, and many more.

Does Adrian offer real-time security monitoring for AI agents?

Yes, Adrian offers real-time security monitoring for AI agents. It is consistently vigilant about the operations and behaviours of AI agents, promptly catching and dealing with events such as misuse of tools, prompt injections, and policy drifts. This allows Adrian to tackle any potentially harmful or misaligned behaviour before the AI agent goes ahead with the action.

How does the Audit mode in Adrian differ from its Block mode?

Adrian has two fundamental operational modes: the audit mode and block mode which operate differently for enhanced AI agent security. The Audit mode specializes in identifying and classifying security events. It essentially monitors the operations and raises red flags when it spots suspicious activities. The Block mode, on the other hand, actively disrupts these detected dubious activities, providing active security control measures on top of monitoring.

What is Adrian?

Adrian is an open-source runtime security monitoring and control tool designed specifically for Artificial Intelligence (AI) agents.

How does Adrian provide security for AI agents?

Adrian provides security for AI agents by continually monitoring their activities, such as tool calls, actions, and outputs. Apart from monitoring agent activity logs, it also examines reasoning traces to identify and counteract any harmful or misaligned behaviors. This real-time analysis allows Adrian to catch and handle events such as malicious tool use, prompt injection, and policy drift before the AI agent acts.

What kind of events can Adrian catch and handle?

Adrian is capable of detecting and handling events like malicious tool use, prompt injection, and policy drift in real-time, before the AI agent acts.

Can Adrian stop an AI agent's actions?

Yes, Adrian has the capability to intervene and halt an AI agent's suspect activities. This is especially true in its block mode of operation, where it interrupts the agent's actions if they are deemed suspicious.

In what ways can Adrian be integrated into LangChain agents?

Adrian can be integrated into LangChain agents through a Python Software Development Kit (SDK). With just two lines of code, developers and AI practitioners can install and integrate Adrian with LangChain agents.

What are the two primary operation modes of Adrian?

Adrian operates in two primary modes: audit mode and block mode.

What is the difference between the audit mode and block mode in Adrian?

In audit mode, Adrian identifies and classifies security events for further analysis. However, in block mode, Adrian takes an additional step and interrupts any identified suspect activities.

What are the key features of Adrian?

The key features of Adrian include its ability to alert via multiple channels, the capacity to delineate acceptable behaviours from known risks, compatibility for self-hosting on a user's own infrastructure, the ability to analyse agent activity logs and reasoning traces, real-time detection and response to harmful behaviours, and in-flight intervention capabilities.

Is Adrian capable of self-hosting on a user's infrastructure?

Yes, Adrian has the capability to be self-hosted on a user's own infrastructure. This makes it a highly flexible solution suitable for a variety of use cases that require real-time AI security monitoring and control.

Can Adrian be used for real-time AI security monitoring and control?

Yes, Adrian is built to offer real-time AI security monitoring and control.

Is Adrian open-source?

Yes, Adrian is an open-source tool.

Where can I find Adrian on GitHub?

Adrian can be found in the secureagentics repository on GitHub.

How does Adrian analyze AI behavior?

Adrian analyses AI behaviour by inspecting agent activity logs, including tool calls, outputs, actions, and the agent's reasoning traces. This allows it to detect any deviant behaviours and react in real-time, before the agent acts.

What does Adrian provide for developers and AI practitioners?

Adrian offers developers and AI practitioners a straightforward and accessible way to ensure their AI tools and models operate within their specified limits. It helps reduce the risk of unexpected or unwanted behaviors by providing real-time monitoring and control capabilities.

Can Adrian intervene in-flight for deviant behaviours?

Yes, Adrian allows for in-flight intervention on behaviours that deviate from expected or acceptable patterns, adding an extra layer of security during the runtime operation of AI agents.

Is there an SDK for Adrian?

Yes, there is a Python Software Development Kit (SDK) for Adrian, which can be easily installed and integrated.

Does Adrian provide an alert system?

Yes, Adrian provides an alert system. Users can set up specific alerting channels as part of the tool's customizable features.

Can Adrian be customized in regards to defining acceptable behaviours and known risks?

Absolutely, Adrian can be customized to define acceptable behaviours and known risks. This allows users to have precise control over the operation of their AI agents.

How does Adrian help ensure AI policy compliance?

Adrian helps ensure AI policy compliance by continuously monitoring the behaviour of AI agents in real-time. Any anomalies, such as policy drift or deviant behaviour, are detected before the AI agent executes its actions. This also extends to in-flight intervention for behaviours that deviate from the accepted norm.

What is the role of Adrian in threat detection and prevention?

Adrian plays a crucial role in threat detection and prevention. By scrutinising an agent's activity logs and reasoning traces, it identifies deviant behaviours and reacts in real time, negating the potential risks before the agent acts. This includes handling events like malicious tool use and prompt injection, thereby adding another layer of security to the AI system.

Ask a question

Submit

#675 2 0

Search

Adrian

Overview

Supported features

Releases

Max Corbridge

Pricing

Top alternatives

Related topics

Reviews

How would you rate Adrian?

Prompts & Results

Pros and Cons

Pros

View 32 more pros

Cons

View 3 more cons

Q&A

Go to section

Search

Overview

Supported features

Releases

Max Corbridge

Pricing

Top alternatives

Related topics

Reviews

How would you rate Adrian?

Prompts & Results

Pros and Cons

Pros

View 32 more pros

Cons

View 3 more cons

Q&A

Help

People also viewed

Feedback and Incident Report

AI Options

Create AI Tools

Mini Tool

Vibe code an AI Tool