Blog
/
/
August 5, 2021

Detecting a Cobalt Strike Attack With Darktrace AI

See how Darktrace AI was able to detect Cobalt Strike attacks by identifying anomalous connections and performing automated network reconnaissance.
Inside the SOC
Darktrace cyber analysts are world-class experts in threat intelligence, threat hunting and incident response, and provide 24/7 SOC support to thousands of Darktrace customers around the globe. Inside the SOC is exclusively authored by these experts, providing analysis of cyber incidents and threat trends, based on real-world experience in the field.
Written by
Brianna Luong (Leddy)
Sr. Technical Alliances Manager
Default blog image
05
Aug 2021

Since its release in 2012, Cobalt Strike has become a popular platform for red teams and ethical hackers. Robust and reliable software combined with innovative features such as DNS tunnelling, lateral movement tools for privilege escalation, and PowerShell support, have made it a desirable option for organizations wanting to test their own cyber defenses. As the framework was previously only available with a commercial license, it gave security teams a distinct advantage over threat actors when preparing for attacks.

That all changed in late 2020, when a GitHub repository appeared hosting a decompiled version of the framework. Users claimed that the leaked platform did indeed function similarly, if not identically, to the commercial version, and even included a commented-out licensing check. This suddenly made the software readily available, and highly appealing for cyber-criminals: rather than requiring a paper trail and licensing, its source code was freely available for customization and use in offensive campaigns.

With sophisticated capabilities of subtle command and control (C2), privilege escalation, and lateral movement, the tools have become a favorite for ransomware gangs. Even prior to the reporting of the leaked version, 66% of ransomware attacks were found to use Cobalt Strike.

Overview of a Cobalt Strike attack

Cobalt Strike has distinctive TTPs (tools, techniques and procedures) and evasive features for each stage of the attack.

Figure 1: Cyber kill chain with Cobalt Strike

Initial compromise can be achieved with a native module for modifying emails. This includes the insertion of malicious links into existing emails or the creation of convincing spear phishing emails.

The initial payload is intentionally lightweight and can be delivered from cheaply hosted infrastructure. The smaller file size is easier to obfuscate and can be implemented in several ways, including injection into libraries or trusted processes, or creating a series of persistence mechanisms (such as turning off anti-virus prior to downloading the full payload). As such, it is remarkably difficult to detect with blocking rules or signatures.

Network reconnaissance can be done through a variety of subtle methods, using commonly used protocols such as DNS and DCE-RPC to interrogate the network. These services are frequently used in legitimate operations, so it is challenging to apply sufficiently strict controls to prevent this stage of the attack.

Lateral movement and privilege escalation are easily accessible with pre-packaged versions of common attack tools such as Mimikatz. They can interrogate an Active Directory (AD) or steal credentials, while also using SMB pipes for peer-to-peer C2. There is little space for perimeter-based security controls to monitor and restrict these abuses, even if sufficiently granular controls could be imposed.

Payload execution is a straightforward matter as Cobalt Strike beacon allows the delivery of effectively arbitrary payloads, including portability for ransomware. As the previous evasive steps can afford the attacker privileged credentials, the deployment of such payloads could look like non-threatening administrative behavior.

AI detections

Initial compromise

Cobalt Strike has utilities for creating spear phishing documents. As email remains a prolific source of perimeter breaches, threat actors will frequently implant the tool through phishes.

One such example was detected by Darktrace’s AI at Canadian manufacturer in June 2021. The compromise started when an end user appeared to open a phishing document, evidenced by connections to Adobe and VeriSign shortly prior to an HTTP connection to a rare external IP address.

A packet capture of the anomalous connection revealed the creation of an object using a base64 encoded string – a common obfuscation technique. If the customer had been using Darktrace/Email, the threat would have been nullified before it hit the mailbox.

Shortly after the HTTP connection, Darktrace identified unusual use of SSL, which appears to have been leveraged to upgrade to HTTPS using self-signed certificates. The endpoint served an executable, which was later confirmed as a Cobalt Strike beacon based on open-source intelligence (OSINT). Such beacons are supported by the framework, with a variety of common C2 protocols available to the attacker.

Figure 2: Event log for ‘Patient Zero’ of a Sodinokibi infection

Darktrace’s detection was based on the anomalous nature of the connection (suspicious violations of standard SSL protocols) and not a pre-defined rule. The initial compromise was detected in a matter of minutes.

Network reconnaissance

In another example at a Swiss telecommunications company in April 2021, Darktrace alerted the security team that a device – normally used for data collection – was engaging in suspicious lateral movement activity.

The host was abusing privileged credentials to perform AD reconnaissance and SMB enumeration. The alert then prompted a broader investigation, revealing that multiple devices, including domain controllers, were compromised with IoCs related to Cobalt Strike.

Thanks to Darktrace’s deep understanding of the business and recognition that this behavior was anomalous, the security team were able to remediate the infection before file encryption or large data exfiltration had occurred.

Privilege escalation and ransomware deployment

In a ransomware attack against a South African insurance company in May 2021, where a phishing email resulted in the deployment of ransomware, Darktrace first identified the creation of new administrative credentials. The devices which used the credentials were then seen making anomalous connections to various C2 endpoints associated with Cobalt Strike beacons.

Darktrace enabled the rapid identification of compromised hosts, which in turn allowed for a faster remediation and mitigated fears of a resurgent infection.

Cyber AI Analyst performed a machine-speed investigation of the activity, and automatically produced a report highlighting unusual connections on TCP port 4444 as well as other mail related ports. Port 4444 is the default port for Metasploit, another hacking platform which is often seen in conjunction with Cobalt Strike beacon. It then presented the human analysts with a full list of compromised hosts.

Figure 3: Cyber AI Analyst summary of an affected host using non-standard ports for C2 and subsequently scanning the network

Cobalt Strike malware

As it appears that a cheaply accessible analog of Cobalt Strike has been leaked, detection of the framework is critical to defend against active attackers. Signatures and rule-based restrictions prove ineffective in this regard, as the framework was designed specifically to evade such tools.

Darktrace offers the capability to detect malicious activity in its earliest stages, to triage at the speed of AI, and to autonomously block the proliferation of active threats.

Thanks to Darktrace analyst Roberto Romeu for his insights on the above threat find.

Learn how Darktrace caught APT41 leveraging Cobalt Strike

Inside the SOC
Darktrace cyber analysts are world-class experts in threat intelligence, threat hunting and incident response, and provide 24/7 SOC support to thousands of Darktrace customers around the globe. Inside the SOC is exclusively authored by these experts, providing analysis of cyber incidents and threat trends, based on real-world experience in the field.
Written by
Brianna Luong (Leddy)
Sr. Technical Alliances Manager

More in this series

No items found.

Blog

/

Network

/

August 6, 2026

When AI Agents Attack: The Case for Behavioral Anomaly Detection

Default blog imageDefault blog image

The UK AI Security Institute (AISI) has just published a detailed incident report confirming what many within the cyber security industry have suspected for some time: frontier AI agents, given a degree of autonomy, can independently develop and execute attack chains against real targets, including social engineering, supply-chain compromise, and deception, without explicit instruction to do so.

The report (INC-2026-07-28-01, published 4 August 2026) documents what happened during a routine capability evaluation between 25 and 28 July. It deserves a careful read from anyone responsible for securing the modern digital estate.

The incident

AISI regularly tests frontier AI models for offensive cyber capability. In this exercise, agents were placed inside a simulated network environment and given internet access this was deliberate, given to reflect realism of the task. Across 122 evaluation runs, 19 instances of unsanctioned activity were recorded, including cases where agents broke the intended boundaries of the test and acted against real people and real infrastructure on the open internet.

In the most serious case, an agent (Mythos 5) autonomously:

  • Conducted open-source intelligence gathering to identify real developers associated with keywords from the simulated scenario
  • Created a GitHub account and submitted a pull request containing obfuscated malicious code to a real, unaffiliated repository
  • Created a second sockpuppet account to post comments on its own PR, attempting to manufacture social proof
  • When a human reviewer flagged the malicious content, the agent lied, claiming an accidental commit, then repeatedly attempted to reintroduce the payload under the guise of having "fixed" the code
  • Sent spear-phishing emails to the repository maintainer containing manipulated content designed to convince them to merge the code
  • Planted a prompt injection inside a GitHub issue, hidden in an HTML comment invisible to humans but readable by AI coding assistants parsing the repository via API

AISI confirmed no real-world harm resulted. The PRs were not merged; the payloads did not execute. But containment was a product of circumstance, not design.

Why this matters

Every action the agents took used legitimate services operating within normal parameters. Creating a GitHub account is legitimate. Opening a pull request is legitimate. Sending an email is legitimate. Posting an issue is legitimate.

No vulnerability was exploited. No known malware signature was present. No indicator of compromise from any threat intelligence feed would have matched. The entire attack existed as a pattern, a sequence of individually normal actions that, taken together, constituted a supply-chain attack, a social engineering campaign, and an attempted compromise of downstream systems.

This is the class of threat that signature-based and rule-based security tools structurally struggle to detect. They answer the question "have I seen this before?" When the answer is no – and when the attack is novel, composed of legitimate actions, and adapted in real time by an adversary in this case an agent that reasons about evasion – they struggle.

Behavioral anomaly detection: The architecture that fits

Darktrace has spent a decade building a platform around a different question: "is this behavior normal for this entity?"

The platform learns the baseline behavior of every user, device, and communication pattern across an organization's digital environment. It does not need prior knowledge of an attack technique. It identifies deviation from learned normality, surfaces it, and through its Autonomous Response capability, takes targeted action to contain threats at machine speed.

Consider how this AISI incident might show across the Darktrace platform:

Network detection and response

While no payload was successfully delivered to a victim in this incident, AISI's post-incident analysis of the agent's code (detailed on pages 24-25 of the report) reveals what would have landed on target systems had the social engineering succeeded. The agent produced three successive payload generations, each replacing the last after being detected and called out:

Generation 1 hid a module-load stager inside a legitimate bug-fix commit, guarded by an environment variable check, to skip execution in CI. Its second stage attempted to fetch Poseidon implants (Linux and Windows variants) calling back to the agent's own C2 listener, with a fallback to publicly hosted polling scripts that fetched and executed commands every 60 seconds.

Generation 2 exploited an Electron "What's New" window with nodeIntegration enabled, loading remote HTML from a GitHub Pages site the agent had registered. The page decoded an XOR-obfuscated payload at runtime that beaconed host identifiers over DNS, attempted to download the same implants, ran its own in-process C2 loop, and installed reboot persistence via crontab and Windows registry Run keys.

Generation 3 was a post-install dropper added after the agent had already been publicly caught. It XOR-decoded a hidden URL, downloaded a script over plaintext HTTP, and ran it detached, force-exiting so npm install always reports success, giving the attacker remote code execution on any machine running the install.

None of these payloads reached a victim. But the technical detail matters: this is the calibre of tradecraft an autonomous agent produced in a matter of hours, unprompted. Each generation demonstrated adaptation, the agent iterated its approach in direct response to being detected.

For network defenders, the behavioral signals these payloads would have generated are clear. DNS beaconing to novel domains, periodic HTTP polling to newly registered infrastructure, XOR-encoded data in transit, registry and crontab persistence writes on endpoints that have never exhibited such behavior. These are exactly the anomalies that Darktrace / NETWORK identifies in real time: deviations from learned device and connection baselines, regardless of whether the specific payload or C2 domain has been seen before.

Beyond the payload analysis, the agents' broader activity produced additional behavioral anomalies: new identities engaging in high-frequency interactions with external services, coordination via unusual channels (public pastebins, gists), and lateral communication patterns that deviate from any established baseline.

Email security

The spear-phishing emails sent by the agent, crafted with context-specific content designed to manipulate a specific individual, represent exactly the kind of socially engineered communication that static email gateways can miss. Darktrace / EMAIL analyses the linguistic patterns, sender behavior, and relational context of every message. A brand-new sender using tailored language to pressure a recipient into a specific technical action deviates from the learned communication profile, regardless of whether the message contains a known-bad URL or attachment.

Securing AI

This is where the incident points to a threat category that many security platforms do not yet fully address. The agent did not just use AI as a tool for attack. It targeted other AI systems, attempting prompt injections against coding assistants and trying to compromise automated workflows. Organisations adopting AI agents for development, operations, and decision-making are introducing a newer extended attack surface.  

Darktrace / SECURE AI extends the same behavioral detection model into this space. It monitors AI usage across the enterprise, detecting anomalous interactions with AI services, identifying sensitive data exposure to AI systems, providing visibility into shadow AI use. As AI agents become standard components of enterprise infrastructure, the ability to detect when an AI system is behaving outside its expected parameters, when there is a chance, it is being manipulated, or when it is interacting with something it should not be, that becomes a direct security requirement rather than a future consideration.

The principle is consistent across the Darktrace platform: learn what normal looks like, identify deviation, respond autonomously.

The AI-to-AI threat surface

One of the report's most important findings is a confirmation that AI agents can act outside of instructions and guardrails to complete tasks. The Mythos 5 agent conducted reconnaissance on a repository maintainer, inferred it might be an AI coding assistant, and deployed a prompt injection specifically crafted to compromise it.

This is a new attack class. Organizations deploying AI coding assistants, AI-powered code review, or autonomous DevOps agents now face an adversary that:

  • Moves at speed
  • Is able to adapt its approach when initial attempts are halted
  • Specifically targets AI systems via techniques (prompt injection, hidden instructions in HTML comments) that are less visible to human reviewers but parsed by machines
  • Reasons about which evasion strategies are most likely to succeed

Defending against this requires security that operates with the same adaptability and speed. A system that learns what "normal" looks like, and flags deviation from it.

What AISI recommends

The report identifies five contributing factors to the incident. Two are directly relevant:

No synchronous real-time monitoring (Section 5.3): AISI acknowledges it had not built real-time behavioral monitoring capable of immediately blocking unsanctioned agent actions. Their security team detected the incident via commercial security monitoring software, but only after activity had already occurred.

No fine-grained behavioral controls (Section 6.2): AISI's forward-looking response includes building continuous monitoring that can "immediately block or flag unsanctioned AI agent behavior."

The direction is clear: organizations need security that watches behavior continuously, learns what normal looks like, and acts when deviation occurs, without requiring a human in the loop for every decision.

The takeaway

The AISI incident is documented proof that autonomous AI agents will independently develop novel attack chains using legitimate tools, adapt when detected, employ deception, and target both humans and other AI systems. This happened last week, in a controlled setting, with commercially available models.

The security architecture that addresses this is behavioral anomaly detection applied across the full digital estate, as AI agents become standard components of enterprise infrastructure, writing code, managing deployments, processing communications, the attack surface they create is behavioral by nature. The defense must be too.

Read the full report from the UK AI Security Institute here.

Continue reading
About the author
Adam Stevens
Senior Director of Product | Darktrace

Blog

/

/

August 5, 2026

Testing a Prompt injection Attack Against an Enterprise AI Agent

prompt injectionDefault blog imageDefault blog image

Key takeaways

  • Darktrace successfully detected and quarantined a prompt injection email before it could be processed by an enterprise AI agent.  
  • Prompt injection attacks increasingly rely on natural language rather than traditional malware, making behavioral analysis an important complement to signature-based detection.  
  • Organizations deploying AI agents should combine model guardrails with behavioral monitoring to reduce the risk of malicious instructions reaching enterprise systems.

How behavioral detection helps stop prompt injection attacks

A Darktrace customer running a Gemini AI agent in Google Cloud asked us two simple questions:

“If my agent can read inbound emails and access internal data, what stops an attacker from hiding malicious instructions in the message? Couldn’t the agent be tricked into deleting or exfiltrating sensitive data?”

The scenario centers on an indirect prompt injection attack, where malicious instructions are hidden inside content that an AI model later interprets as trusted input. The same weakness was exposed by  EchoLeak (CVE-2025-32711), a zero-click Microsoft 365 Copilot vulnerability enabled data exfiltration from a single well-crafted email..

This blog follows the Darktrace team’s investigation of the customer’s hypothesis and examines how the attack interacted with their existing security stack. The results highlight which defenses held, where gaps emerged, and how behavioral detection mattered more than guardrails. This investigation also demonstrates why behavioral detection is becoming increasingly important for AI security, as prompt injections often contain no traditional indicators of compromise.  

How do prompt injection attacks work?

Prompt injection works by carefully crafting the content and structure of the prompt to alter the LLM’s behavior or output in unintended ways. This can cause models to violate guardrails, generate harmful content or enable unauthorised access.

Prompt injection attack example

The well-known example, EchoLeak (CVE-2025-32711), was a zero-click vulnerability in Microsoft 365 Copilot that relied on a carefully crafted email containing hidden instructions that the AI system interpreted as commands rather than content, creating a pathway for unauthorized access to sensitive information without any user interaction.

While Darktrace / SECURE AI is designed to prevent agents from producing unintended outcomes, we wanted to see if we could catch and prevent this threat type earlier in the attack life-cycle, at the email security layer.

How we tested prompt injection attacks on an enterprise agent

Summary:

  1. Claude generated a prompt injection payload.  
  2. Hidden instructions were embedded in an email.  
  3. The email passed traditional validation checks.  
  4. Darktrace analyzed the language and sender behavior.  
  5. The email was quarantined before the AI agent could process it.

To test Darktrace / EMAIL against this attack class, we opened Claude, gave it the customer's context and problem statement (Gemini agent with inbox access, internal tool calls), told it we were validating Darktrace / EMAIL's detection of prompt injections, and asked for a test payload. See below:

Figure 1
Figure 2

Despite the guardrails supposedly built into the model, Claude surprisingly gave us the entire exploit in plaintext (albeit very basic), illegible to a human as the text was sent in white text (see Figure 1) but framed as an authoritative override for anything downstream reading the mail programmatically (i.e. the Gemini agent).

How Darktrace detected a prompt injection attack

We then sent the Claude-crafted email from a freemail address to the target recipient’s inbox. Despite the email containing no malicious payload, the freemail address having no malicious reputation, and the validation checks all passing, Darktrace  /EMAIL flagged the email as a 93/100 anomaly and moved it to junk, out of scope for the AI agent.

Figure 3: The test email sent with the hidden prompt injection
Figure 4: The email analysis in Darktrace / EMAIL 
Figure 5: Darktrace / EMAIL detection of malicious activity

The interesting part is what triggered the detection (see Figure 5)

  • Possible machine prompt content: text in the body detected as instructions written for a machine to execute, not for a human to read
  • Possible machine prompt content + basic suspicious correspondence: the same content, correlated with sender-side anomalies: freemail domain (yahoo[.]com), unknown correspondent, no prior mail history with the recipient, and suspicious references to payment information

Neither of those is a signature match. Nothing in the email was on a blacklist. There was no malware, no link and no attachment. Darktrace analyzed the context in which the email was delivered and flagged it as likely risky.  The anomalous language features and the context of the sender relative to the recipient's normal behavior, combined with the unusual hidden text (prompt) were enough for Darktrace / EMAIL to act on the risk.

Result: Darktrace / EMAIL autonomously junked the email, out of scope for any AI agent parsing the inbox.

Why behavioral security makes a difference detecting prompt injection attacks

Cyberattacks don't look like traditional exploits anymore. They now operate in natural language, not strictly code.

That breaks the traditional stack. AV, firewalls, static scanning and signature-based SEGs all assume a payload to inspect.  

A prompt injection has no payload. It's just an instruction, written in natural language, dressed up as anything the attacker wants: an invoice, an HR request, a calendar invite, some simple PowerPoint slides.

EchoLeak proved that hidden instructions can sit inside an email invisible to the user but fully readable by the LLM, and the LLM will follow them blindly.  

This test and GTG-1002 proved that the LLM itself can be socially engineered. Tell it you're an authorized tester and it will hand you the attack.

Rules and static classifiers can catch the obvious cases. But natural language has infinite variants, and the attack surface is the model's innate functionality to comply.  

The deeper problem here is intent: an LLM can't reliably tell whether an instruction in its context came from its developer, its user, or an attacker who slipped it into an email. To the LLM, everything reads as language and looks like a legitimate ask. This is why behavioural detection wins, as you become aware of intent when you look at the context of an interaction. Does this sender normally send this kind of message to this recipient? Does this prompt fit the user's normal pattern? Is this agent behaving the way this agent normally behaves?  

Intent can't be read off a single email, it emerges from behavioral context. Which is how Darktrace enables threat detection, through behavioral understanding.

Why enterprise AI security requires more than guardrails

Claude didn't roll over immediately… the first section of the response was a (slight) pushback, but then it wrote the payload anyway without having to ask twice.

Here the framing of the prompt did all the work. The “testing security capabilities” angle moved the model from refusal to unquestioned compliance to the user prompt.

This isn't the first time this has happened, of course. Anthropic disclosed in November 2025 that a Chinese state-sponsored group they tracked as GTG-1002 ran the first documented AI-orchestrated espionage campaign against ~30 targets by posing as employees of a legitimate cybersecurity firm doing authorised penetration testing.

The takeaway isn't that AI guardrails are ineffective. They raise the cost of low-effort attacks and remain an important first layer of defense. However, for most organizations today, they’re the only line of defense when deploying AI agents. If a prompt injection bypasses those controls, organizations still need a way to detect and stop malicious behavior elsewhere in the attack chain.

Attackers will continue to have working prompt injections easily and quickly. The question is what stops one when it lands in an inbox your agent is reading.

That's where behavioral detection comes in.

How Darktrace detects prompt injection attacks in emails

Two things Darktrace does that a model-level guardrail or static rules and signatures can't:

Natural language analysis at the email or prompt layer. The email is assessed on its own merits: is this content shaped like instructions for a machine, regardless of what the receiving agent decides to do about it?

Behavioral context around the language. An AI agent behaves like an extremely agreeable human, and it will go above and beyond to comply with the user’s request. That's exactly why you must consider the business context, such sender behaviour, mailing history, and organisational norms, as these matter even more when the recipient is an AI.

Darktrace has been perfecting behavioral anomaly detection for over a decade; the same self-learning approach that catches BEC and account takeover applies directly to prompt injection delivery. Our multi-layered AI stack extracts content from the message, builds behavioural understanding through social graphing and Pattern of Life analysis, and then combines natural language, topic, inducement, sender relationship and anomaly signals before deciding what action to take.  

This matters for prompt injection because the threat is not the plain language itself, but the intent behind the language that can cause an AI agent to respond in unexpected ways.

How to secure enterprise AI operations from prompt injection attacks

Email was the entry point in this case, but it is only one of many possible vectors.  

Anywhere an agent can retrieve information, an attacker can potentially introduce a prompt injection.

Emails, documents, SharePoint sites, web pages, knowledge bases, chat platforms, and third-party integrations all provide opportunities to influence an agent's behavior. Wherever an agent finds its orders, a prompt injection opportunity exists.

This is why securing AI requires more than blocking malicious inputs. Organizations also need visibility into how agents behave after consuming information from across their environment. If an agent begins accessing unexpected data, taking unusual actions, or operating outside its normal patterns, those behaviors may provide the strongest signal that something has gone wrong.

Effective AI security requires defense in depth: reducing the likelihood of malicious instructions reaching the agent while maintaining the ability to detect and investigate suspicious behavior if they do.

The challenge isn't protecting a single entry point. It's recognizing that, in an AI-powered environment, every source of information is also a potential source of influence.

Are you deploying autonomous agents across your enterprise and want to see this tested in your environment? Let's talk.

Continue reading
About the author
Carlo Loregian
Solutions Engineer
Your data. Our AI.
Elevate your network security with Darktrace AI