Outlook’s cozy reputation takes a hit as hackers slip in through a sneaky add-in first seen harming labs. What makes this case more alarming is that it involves a Malicious Outlook Add-In hiding inside a trusted system. Trusted features turn into weapons when a calendar tool named AgreeTo siphons logins without warning.
Researchers tag the campaign “AgreeToSteal,” watching silently as credentials vanish into the shadows. Four thousand accounts have already been touched, and numbers may continue climbing without loud alarms. It is not just faulty code at fault. Faith in app stores begins to crack under quiet infiltration, as one Malicious Outlook Add-In slowly unravels trust piece by piece.
Outlook by Microsoft ranks among the top email platforms globally. Every single day, countless individuals rely on it for job-related messages, classroom updates, lessons, and private notes. Built-in extras, such as add-ins, can be added straight into the system to boost productivity—but a single Malicious Outlook Add-In can turn a helpful tool into a serious security threat.
Instead of doing everything manually, tools handle meeting times, track tasks, and sort incoming mail neatly. People naturally trust tools found inside the official Microsoft Office Add-ins Store. That strong belief in safety opened the door for this attack to succeed.
A small tool named AgreeTo played a key role in the incident. Released near the end of 2022, it started as a meeting planning helper. Users could share calendar availability quickly and schedule discussions with ease. Its original goal was simple: make daily work smoother.
However, updates eventually stopped. The developer walked away from the project. Few people noticed when the developer’s website domain expired. Even after that, the add-in remained listed in Microsoft’s online store.
Outlook add-ins are not traditional software programs installed on a computer. They function more like web pages loaded inside Outlook. Developers submit a special document called a manifest, written in XML, when creating one.
This file instructs Outlook to open a specific web address inside an iframe. Microsoft reviews the manifest and approves it before listing the add-in in its store. But here is the key detail: every time the add-in opens, it pulls fresh content directly from the developer’s server.
That detail created the weakness. When the original domain expired, attackers registered it. With control over the web address, they replaced the real calendar scheduling page with a fake Microsoft login form. When users opened the AgreeTo extension, they no longer saw a meeting planner. Instead, they saw what looked like an authentic Microsoft sign-in screen. Because it appeared inside Outlook, many users did not question it.
Once victims entered their email and password, a simple JavaScript function captured the typed information. Their IP address was quietly collected at the same time. The data was then sent directly to attackers using Telegram’s Bot API. No complex infrastructure was needed. A single fetch request handled the transfer. The attack did not rely on advanced malware. Its strength came from placement and trust.
After typing their credentials, users briefly saw a loading spinner. Then they were redirected to the real Microsoft login page. Thinking the first attempt had simply failed, they logged in again. The second login worked normally, giving them access to their account. Most victims never realized their credentials had already been stolen. They continued using Outlook while attackers potentially monitored their messages.
Security experts explain that the phishing design itself was basic. The power of the attack came from distribution through Microsoft’s official add-in marketplace. A familiar permissions prompt added to the sense of legitimacy. When something appears inside a trusted storefront and interface, doubt fades quickly. Users assume safety because the environment feels official.
The AgreeToSteal operation may have exposed more than four thousand Microsoft accounts. These accounts could contain personal conversations, business discussions, financial information, and sensitive documents. With access to a single inbox, attackers can reset passwords for other services, send phishing emails to contacts, or search messages for valuable information.
Outlook add-ins can request permissions such as ReadWriteItem access. This allows them to read or modify message content. If harmful scripts operate under these permissions, the damage could go beyond stealing passwords. Attackers could read private emails, alter communications, or send messages pretending to be the victim. The potential risk grows significantly under such access levels.
This incident highlights a broader supply chain security issue. Instead of attacking users directly, hackers targeted a trusted platform. By hijacking a previously legitimate tool, they reached thousands of users at once. Microsoft reviews add-ins before listing them, but this case shows that risks can appear later if developers fail to secure their domains and servers.
Security specialists warn developers to monitor their domain names carefully. When a domain expires, anyone can register it. If that domain is linked to a trusted tool, attackers may misuse it. That is how a once helpful extension transformed into a Malicious Outlook Add-In without changing its listing in the store.
Users must also remain alert. Even on trusted platforms, unusual behavior should raise concern. If an add-in suddenly displays an unexpected login screen, it could be suspicious. Always check the website address before entering credentials. If something seems unusual, close the window and sign in directly through Microsoft’s official website.
For companies, this event serves as a strong reminder to enable multi-factor authentication. A second verification step, such as a code sent to a mobile device, can prevent unauthorized access even if passwords are stolen. Organizations should also review installed add-ins regularly and remove those that are no longer necessary.
The misuse of Telegram’s Bot API in this attack shows how everyday tools can be exploited. Telegram is widely used for normal communication, but attackers leveraged its automation features to collect stolen data quickly. Technology itself is neutral. Its impact depends on how it is used.
This marks the first known real-world case involving a harmful Outlook add-in in active use. It signals that cybercriminals are exploring new methods. Instead of creating complex viruses, they may exploit trusted ecosystems and simple scripts. By blending into everyday workflows, they increase their success rate.
Email remains one of the most important communication tools in the world. Protecting it requires awareness from developers, platform providers, and users alike. Add-ins can improve productivity, but they must be carefully monitored and maintained.
The key lesson is clear. Digital trust must be protected continuously. Abandoned projects and expired domains can create hidden entry points for attackers. By learning from incidents like AgreeToSteal, the technology community can strengthen defenses and reduce future risks.
As the internet grows, so do cyber threats. Staying cautious, using strong passwords, enabling multi-factor authentication, and paying attention to unusual activity remain the best defenses. Smart habits and informed decisions protect users far better than blind trust ever can.
