salesforce.com Eoghan Casey
August 27, 2025
Learn how to detect, investigate, and respond to Salesforce security incidents with logs, permissions, and backups.
A guide to investigating Salesforce security incidents with logs, permissions, and backups to strengthen response and resilience.
I am increasingly asked by customers how to investigate potential security incidents in their Salesforce environments. Common questions are: What did a specific user do during that time? and What data was impacted? Every organization and incident is unique, and the answer to these questions depends on the specific situation, but there is some general guidance I can provide.
Three key sources of information for investigating a security incident in Salesforce environments are activity logs, user permissions, and backup data.
bbc.com Chris VallanceSenior Technology Reporter andTheo Leggett International Business Correspondent 3.09.2025
Staff were sent home and the company shut down its IT systems in an effort to minimise the damage done.
A cyber-attack has "severely disrupted" Jaguar Land Rover (JLR) vehicle production, including at its two main UK plants.
The company, which is owned by India's Tata Motors, said it took immediate action to lessen the impact of the hack and is working quickly to restart operations.
JLR's retail business has also been badly hit at a traditionally a popular time for consumers to take delivery of a new vehicle - but there is no evidence any customer data had been stolen, it said.
The attack began on Sunday as the latest batch of new registration plates became available on Monday, 1 September.
The BBC understands that the attack was detected while in progress, and the company shut down its IT systems in an effort to minimise any damage.
Workers at the company's Halewood plant in Merseyside were told by email early on Monday morning not to come into work while others were sent home, as first reported by the Liverpool Echo.
The BBC understands the attack has also hit JLR's other main UK manufacturing plant at Solihull, with staff there also sent home.
The company said: "We took immediate action to mitigate its impact by proactively shutting down our systems. We are now working at pace to restart our global applications in a controlled manner."
It added: "At this stage there is no evidence any customer data has been stolen but our retail and production activities have been severely disrupted."
It is not yet known who is responsible for the hack, but it follows crippling attacks on prominent UK retail businesses including Marks & Spencer and the Co-op.
In both cases, the hackers sought to extort money.
While JLR's statement makes no mention of a cyber-attack, a separate filing by parent company Tata Motors to the Bombay Stock Exchange referred to an "IT security incidence" causing "global" issues.
The National Crime Agency said: "We are aware of an incident impacting Jaguar Land Rover and are working with partners to better understand its impact."
In 2023, as part of an effort to "accelerate digital transformation across its business", JLR signed a five-year, £800m deal with corporate stablemate Tata Consultancy Services to provide cybersecurity and a range of other IT services.
The halt in production is a fresh blow to the firm which recently revealed a slump in profits attributed to increasing in costs caused by US tariffs.
helpnetsecurity.com Zeljka Zorz, Editor-in-Chief, Help Net Security
September 2, 2025
Zscaler, Palo Alto Networks, PagerDuty, Tanium, and SpyCloud say their Salesforce instances were accessed following the Salesloft breach.
The companies noted that attackers had only limited access to Salesforce databases, not to other systems or resources. They warned, however, that the stolen customer data could be used for convincing phishing and social engineering attacks.
The Salesloft breach
Salesloft is the company behind a popular sales engagement platform of the same name.
The company’s Drift application – an AI chat agent – can be integrated with many third-party platforms and tools, including Salesforce.
On August 26, Salesloft stated that from August 8 to August 18, 2025, attackers used compromised OAuth credentials to exfiltrate data from the Salesforce instances of customers that have set up the Drift-Saleforce integration.
Several days later, the Google Threat Intelligence Group (GTIG) confirmed that the compromise impacted other integrations, as well.
“On August 28, 2025, our investigation confirmed that the actor also compromised OAuth tokens for the ‘Drift Email’ integration. On August 9, 2025, a threat actor used these tokens to access email from a very small number of Google Workspace accounts,” GTIG analysts shared.
Astrix Security researchers have confirmed that the attackers used the Drift Email OAuth application for Google Workspace to exfiltrate emails and that – at least in one case – they tried to access S3 buckets whose names have been likely extracted from compromised Salesforce environments.
Similarly, WideField threat researchers have observed suspicious log event activity across multiple customers using its security platform, pointing to attackers rifling through Salesforce databases and Gmail accounts.
Salesloft breach victims Zscaler
How UNC6395 accessed emails (Source: WideField)
Zscaler, Palo Alto Networks and the other companies mentioned above are just some of the 700+ companies impacted by this breach.
While the stolen customer information can be valuable, GTIG analysts say that the attackers were focused on searching for AWS access keys, passwords, and Snowflake-related access tokens, which can (and likely have been) further misused by the attackers.
What to do if your organization is on the victims list?
Salesloft has yet to reveal how the attackers managed to get their hands on the OAuth tokens they used, but the company has engaged cybersecurity experts from (Google’s) Mandiant and Coalition to help them investigate and remediate the compromise.
“We are recommending that all Drift customers who manage their own Drift connections to third-party applications via API key, proactively revoke the existing key and reconnect using a new API key for these applications. This only relates to API key-based Drift integrations. OAuth applications are being handled directly by Salesloft,” the company said on August 27, and outlined the process for updating the API keys.
Salesforce has, for the moment, disabled all integrations between Salesforce and Salesloft technologies, including the Drift app.
“Disabling the connection is a precautionary measure to help safeguard customer environments while we continue to assess and address the situation. We recognize this change may cause disruption and will provide further updates as more information becomes available,” the company noted.
Likewise, Google has disabled the integration functionality between Google Workspace and Salesloft Drift pending further investigation, and has advised organizations to “review all third-party integrations connected to their Drift instance, revoke and rotate credentials for those applications, and investigate all connected systems for signs of unauthorized access.”
Google Mandiant incident responders have provided extensive advice on how organizations can investigate for compromise and scan for exposed secrets and hardcoded credentials.
Astrix researchers have shared additional indicators of compromise and described AWS-specific activity to look out for. WideField threat analysts have provided guidance useful to both their customers and other affected organizations.
blog.checkpoint.com ByAmit Weigman | Office of the CTO September 2, 2025
Researchers analyze Hexstrike-AI, a next-gen AI orchestration framework linking LLMs with 150+ security tools—now repurposed by attackers to weaponize Citrix NetScaler zero-day CVEs in minutes.
Key Findings:
The emergence of Hexstrike-AI now provides the clearest embodiment of that model to date. This tool was designed to be a defender-oriented framework: “a revolutionary AI-powered offensive security framework that combines professional security tools with autonomous AI agents to deliver comprehensive security testing capabilities”, their website reads. In this context, Hexstrike-AI was positioned as a next-generation tool for red teams and security researchers.
But almost immediately after release, malicious actors began discussing how to weaponize it. Within hours, certain underground channels discussed application of the framework to exploit the Citrix NetScaler ADC and Gateway zero-day vulnerabilities disclosed last Tuesday (08/26).
This marks a pivotal moment: a tool designed to strengthen defenses has been claimed to be rapidly repurposed into an engine for exploitation, crystallizing earlier concepts into a widely available platform driving real-world attacks.
Figure 1: Dark web posts discussing HexStrike AI, shortly after its release.
The Architecture of Hexstrike-AI
Hexstrike-AI is not “just another red-team framework.” It represents a fundamental shift in how offensive cyber operations can be conducted. At its heart is an abstraction and orchestration layer that allows AI models like Claude, GPT, and Copilot to autonomously run security tooling without human micromanagement.
Figure 2: HexStrike AI MCP Toolkit.
More specifically, Hexstrike AI introduces MCP Agents, an advanced server that bridges large language models with real-world offensive capabilities. Through this integration, AI agents can autonomously run 150+ cyber security tools spanning penetration testing, vulnerability discovery, bug bounty automation, and security research.
Think of it as the conductor of an orchestra:
The AI orchestration brain interprets operator intent.
The agents (150+ tools) perform specific actions; scanning, exploiting, deploying persistence, exfiltrating data.
The abstraction layer translates vague commands like “exploit NetScaler” into precise, sequenced technical steps that align with the targeted environment.
This mirrors exactly the concept described in our recent blog: an orchestration brain that removes friction, decides which tools to deploy, and adapts dynamically in real time. We analyzed the source code and architecture of Hexstrike-AI and revealed several important aspects of its design:
MCP Orchestration Layer
The framework sets up a FastMCP server that acts as the communication hub between large language models (Claude, GPT, Copilot) and tool functions. Tools are wrapped with MCP decorators, exposing them as callable components that AI agents can invoke. This is the orchestration core; it binds the AI agent to the underlying security tools, so commands can be issued programmatically.
Tool Integration at Scale
Hexstrike-AI incorporates core network discovery and exploitation tools, beginning with Nmap scanning and extending to dozens of other reconnaissance, exploitation, and persistence modules. Each tool is abstracted into a standardized function, making orchestration seamless.
Figure 3: the nmap_scan tool is exposed as an MCP function.
Here, AI agents can call nmap_scan with simple parameters. The abstraction removes the need for an operator to run and parse Nmap manually — orchestration handles execution and results.
Automation and Resilience
The client includes retry logic and recovery handling to keep operations stable, even under failure conditions. This ensures operations continue reliably, a critical feature when chaining scans, exploits, and persistence attempts.
Figure 4: Hexstrike-AI’s automated resilience loop
Intent-to-Execution Translation
High-level commands are abstracted into workflows. The execute_command function demonstrates this. Here, an AI agent provides only a command string, and Hexstrike-AI determines how to execute it, turning intent into precise, repeatable tool actions.
Figure 5: Hexstrike-AI’s execute_command function.
Why This Matters Right Now
The release of Hexstrike-AI would be concerning in any context, because its design makes it extremely attractive to attackers. But its impact is amplified by timing.
Last Tuesday (08/26), Citrix disclosed three zero-day vulnerabilities affecting NetScaler ADC and NetScaler Gateway appliances, as follows:
CVE-2025-7775 – Unauthenticated remote code execution. Already exploited in the wild, with webshells observed on compromised appliances.
CVE-2025-7776 – A memory-handling flaw impacting NetScaler’s core processes. Exploitation not yet confirmed, but high-risk.
CVE-2025-8424 – An access control weakness on management interfaces. Also unconfirmed in the wild but exposes critical control paths.
Exploiting these vulnerabilities is non-trivial. Attackers must understand memory operations, authentication bypasses, and the peculiarities of NetScaler’s architecture. Such work has historically required highly skilled operators and weeks of development.
With Hexstrike-AI, that barrier seems to have collapsed. In underground forums over the 12 hours following the disclosure of the said vulnerabilities, we have observed threat actors discussing the use of Hexstrike-AI to scan for and exploit vulnerable NetScaler instances. Instead of painstaking manual development, AI can now automate reconnaissance, assist with exploit crafting, and facilitate payload delivery for these critical vulnerabilities.
Figure 6: Top Panel: Dark web post claiming to have successfully exploited the latest Citrix CVE’s using HexStrike AI, originally in Russian;
Bottom Panel: Dark web post translated into English using Google Translate add-on.
Certain threat actors have also published vulnerable instances they have been able to scan using the tool, which are now being offered for sale. The implications are profound:
A task that might take a human operator days or weeks can now be initiated in under 10 minutes.
Exploitation can be parallelized at scale, with agents scanning thousands of IPs simultaneously.
Decision-making becomes adaptive; failed exploit attempts can be automatically retried with variations until successful, increasing the overall exploitation yield.
The window between disclosure and mass exploitation shrinks dramatically. CVE-2025-7775 is already being exploited in the wild, and with Hexstrike-AI, the volume of attacks will only increase in the coming days.
Figure 7: Seemingly vulnerable NetScaler instances curated by HexStrike AI.
Action Items for Defenders
The immediate priority is clear: patch and harden affected systems. Citrix has already released fixed builds, and defenders must act without delay. In our technical vulnerability report, we have listed technical measures and actions defenders should take against these CVEs, mostly including hardening authentications, restricting access and threat hunting for the affected webshells.
However, Hexstrike-AI represents a broader paradigm shift, where AI orchestration will increasingly be used to weaponize vulnerabilities quickly and at scale. To defend against this new class of threat, organizations must evolve their defenses accordingly:
Adopt adaptive detection: Static signatures and rules will not suffice. Detection systems must ingest fresh intelligence, learn from ongoing attacks, and adapt dynamically.
Integrate AI-driven defense: Just as attackers are building orchestration layers, defenders must deploy AI systems capable of correlating telemetry, detecting anomalies, and responding autonomously at machine speed.
Shorten patch cycles: When the time-to-exploit is measured in hours, patching cannot be a weeks-long process. Automated patch validation and deployment pipelines are essential.
Threat intelligence fusion: Monitoring dark web discussions and underground chatter is now a critical defensive input. Early signals, such as the chatter around Hexstrike-AI and NetScaler CVEs, provide vital lead time for professionals.
Resilience engineering: Assume compromise. Architect systems with segmentation, least privilege, and robust recovery capabilities so that successful exploitation does not equate to catastrophic impact.
Conclusion
Hexstrike-AI is a watershed moment. What was once a conceptual architecture – a central orchestration brain directing AI agents – has now been embodied in a working tool. And it is already being applied against active zero days.
For defenders, we can only reinforce what has already been said in our last post: urgency in addressing today’s vulnerabilities, and foresight in preparing for a future where AI-driven orchestration is the norm. The sooner the security community adapts, patching faster, detecting smarter, and responding at machine speed, the greater our ability to keep pace in this new era of cyber conflict.
The security community has been warning about the convergence of AI orchestration and offensive tooling, and Hexstrike-AI proves those warnings weren’t theoretical. What seemed like an emerging possibility is now an operational reality, and attackers are wasting no time putting it to use.
clubic.com
Par Alexandre Boero, Journaliste-reporter, responsable de l'actu.
Publié le 01 septembre 2025 à 08h04
La plateforme TikTok Shop commercialise des trackers GPS qui ressemblent au fameux AirTag d'Apple depuis des vidéos virales qui encouragent l'espionnage de ses proches ou de son ou sa partenaire. Les ventes dépasseraient déjà les 100 000 unités.
La marketplace de TikTok héberge des vendeurs de dispositifs de géolocalisation de type AirTag. Les commerçants opèrent leurs ventes à l'aide d'arguments publicitaires qui incitent directement à surveiller secrètement son partenaire. Des vidéos aux millions de vues, des dizaines de milliers de ventes, et une modération défaillante malgré les alertes ont été signalées aux États-Unis. Si la plateforme chinoise affirme interdire ces contenus, elle peine visiblement à les supprimer, ce qui contribue à normaliser les comportements abusifs sur le célèbre réseau social.
Des vidéos à plusieurs millions de vues normalisent sur TikTok l'espionnage conjugal
D'après l'enquête menée récemment par 404 Media, les vendeurs de trackers GPS assument totalement leur positionnement toxique. « Si ta copine dit qu'elle sort juste avec des amies tous les soirs, tu ferais mieux d'en coller un sur sa voiture », peut-on entendre dans une vidéo vue des millions de fois. Le dispositif, carrément présenté comme indétectable contrairement aux AirTags, fait miroiter aux potentiels acheteurs une surveillance mondiale, grâce à la carte SIM intégrée.
Les interactions sous ces publications sont d'ailleurs symptomatiques. Un utilisateur confie dans les commentaires : « J'en ai acheté et les ai mis sur les voitures de filles que je trouve attirantes à la salle de sport. » Oui, c'est flippant, surtout lorsque le vendeur répond avec désinvolture par un émoji rieur. D'après les métriques de TikTok Shop, l'un des traceurs s'est vendu à plus de 32 500 exemplaires, quand un autre affiche quasiment 100 000 unités écoulées.
Eva Galperin, co-fondatrice de la Coalition Against Stalkerware, la coalition contre les logiciels espions, est dépitée. « C'est tout bonnement présenté comme un outil d'abus. » Elle explique que tout dispositif justifié par « attraper son partenaire en train de tromper » facilite le contrôle coercitif. Le pire, c'est que les vidéos multiplient les prétextes pour essayer de toucher plus d'utilisateurs, comme une méfiance conjugale, les références à Coldplay et à l'ex-patron d'Astronomer piégé par une kiss cam, le tout avec des accroches comme « les hommes avec des femmes infidèles, vous pourriez en vouloir un ».
TikTok supprime quelques vidéos mais le problème persiste
Questionné par 404 Media, TikTok a supprimé certaines vidéos et banni un compte, en ajoutant interdire « les contenus qui encouragent la surveillance secrète ». Pourtant, au lendemain de la réponse, le média a déniché des vidéos identiques, qui restaient accessibles. Dès qu'un utilisateur clique sur l'une de ces vidéos, l'algorithme de TikTok Shop lui recommande des produits similaires, notamment des enregistreurs audio secrets vendus avec les mêmes arguments toxiques.
Aux États-Unis, d'où lesdites vidéos ont été publiées, onze États interdisent explicitement le tracking GPS dans leurs lois anti-harcèlement, et quinze considèrent comme illégale la surveillance véhiculaire sans consentement. Les vendeurs jouent sur l'ambiguïté. Certains vont même jusqu'à manier l'ironie dans leur vidéo : « C'est illégal de tracer les gens ? Je ne sais pas, je ne suis pas avocat, mais vous aurez probablement des problèmes ». On n'arrête pas les progrès, mais surtout les dérives.
