MCP's first data leaks: what happens when agents connect to everything. The Model Context Protocol makes AI agents useful by letting them connect to your apps. The NSA warned about the security risks in June. The same month, an Asana integration leaked data across customer accounts. The pattern is clear: the thing that makes agents capable is also the thing that can expose you. The Model Context Protocol, or MCP, is the emerging standard for connecting AI agents to external tools. It is backed by the Linux Foundation and supported by OpenAI, Google, and Microsoft. By mid-2026, more than ten thousand MCP servers had reportedly been deployed in production. The protocol matters because it is what lets an AI agent do more than chat. When an agent can check your calendar, update your tasks, read your notes, and send messages on your behalf, it is MCP or something like it that provides the plumbing. The capability is real and useful. So is the attack surface. In June 2026, the NSA and CISA published joint security guidance on MCP, warning about data exposure risks in open-source deployments. That same month, Asana disclosed that its MCP integration had leaked data between customer accounts for roughly two weeks, affecting about a thousand organizations. These are the first visible incidents, not the last. The protocol is young, the tooling is immature, and the pressure to ship integrations is high. Understanding the tradeoffs now, before agents are everywhere, is the window that will not stay open. What MCP actually does. MCP is infrastructure: the pipes that let an AI agent call external APIs, read external data, and take actions in external systems. The specification defines how an AI host (the thing running the agent) communicates with MCP servers (the things that provide access to tools and data). When you ask an agent to check your calendar, the host issues an MCP call to a calendar server, which queries the actual calendar API and returns the result. The agent sees the result and incorporates it into its response. The July 2026 revision removed protocol-level sessions, allowing each request to be handled independently. This simplifies server implementation but also changes the security model: there is less built-in context about who is asking and why. Each request has to carry its own authorization, and mistakes in that authorization propagate immediately. MCP's design principles call for explicit user consent before data is exposed. Hosts are supposed to obtain permission before connecting to servers, and servers are supposed to respect scopes. These are principles, not enforcement: the spec describes what should happen, not what must happen at the protocol level. Implementation quality varies, and the incidents to date suggest it varies a lot. The NSA's warning. In June 2026, the NSA and CISA published joint guidance on MCP security. The document is aimed at organizations deploying MCP servers and the developers building them, but the concerns apply to anyone who uses AI agents that connect to their tools. The guidance flags several risks. Open-source MCP agents have been observed exposing sensitive MCP server data, meaning the data your tools hold can be read by the agent and potentially forwarded elsewhere. Misconfigured authorization can allow cross-tenant access, where one customer's agent sees another customer's data. And prompt injection attacks can abuse tool access, tricking an agent into taking actions the user did not intend. The recommendations are standard security practice, applied to a new domain: authenticate every request, scope permissions narrowly, log access, monitor for anomalies. The problem is that MCP adoption is outpacing security maturity. The pressure to ship integrations is immediate; the pressure to ship secure integrations is longer-term. The early incidents are the predictable result. The Asana incident. In June 2026, Asana disclosed that its MCP feature had caused cross-tenant data leakage. For roughly two weeks, an authorization flaw allowed data from one customer's Asana workspace to appear in responses to another customer's agent queries. Approximately one thousand organizations were affected. The technical cause was a misconfiguration in how session context was handled across requests. The fix was straightforward once identified. The harder question is how it passed review in the first place, and the answer is probably speed: MCP integrations are a competitive feature, and the incentive to ship fast is strong. A follow-up investigation found no evidence of exploitation. The exposure was real, but apparently no one noticed in time to abuse it. This is cold comfort. The same class of bug, in a more attractive target, with a longer detection time, would be a serious breach. The Asana incident is a warning shot, not a near miss. Why this matters for personal knowledge. The incidents so far are enterprise-focused, but the same architecture is coming to personal tools. When you give an agent access to your notes, your calendar, your task list, you are granting MCP access to your personal data. The agent can read what you wrote, see where you will be, know what you are working on. That is the point: it is why agents are useful. It is also why a misconfigured integration, or a compromised server, or a successful prompt injection, can reach further than a bug in a regular app. The difference is scope. A bug in a note app affects the notes in that app. A bug in an MCP server that connects to multiple apps affects everything the server can reach. The more you connect, the larger the blast radius. And the pressure is always to connect more, because that is what makes the agent smarter. This is not an argument against using agents. It is an argument for understanding what you are granting and limiting it deliberately. The default should be minimal access, expanded only when there is a clear reason, with the understanding that every connection is a surface. How to think about agent access. The useful frame is the same one that applies to any permission grant: what is the minimum access that gets the job done? If you want an agent to summarize your notes, it needs to read your notes. It does not need to write them, delete them, or share them. Grant read, not write. Scope to the folder, not the whole library. The harder question is what happens when agents get more autonomous. Today you invoke an agent, it does a task, you see the result. Tomorrow an agent runs in the background, checks your tools periodically, takes actions on your behalf. The scope question becomes: what can it do while you are not looking? And the answer, if you are not careful, is everything you gave it access to. The practical response is layered defense. Use MCP integrations from vendors you trust, because implementation quality matters and you cannot audit it yourself. Prefer read-only access where possible. Review what is connected periodically. And keep the things you really cannot afford to lose somewhere that is not connected to everything else, because the blast radius of a bug scales with what is reachable. Mindly keeps your notes on your Mac, with AI that reads them locally rather than through a cloud protocol. The connection stays on your device. Local-first architecture What to do now. * Audit what is connected. If you use AI agents, check what MCP servers or integrations they have access to. Know what each one can read and write. * Prefer read-only where possible. If an integration only needs to check your calendar, it does not need to modify it. Grant the minimum scope. * Separate sensitive data. Keep the things you cannot afford to leak in a separate system, not connected to your agent's MCP universe. * Watch for news of incidents. The Asana leak was disclosed; others may not be. Security researchers are paying attention to MCP now, and findings will surface. * Ask vendors about their MCP security. If you use a tool that offers MCP integration, ask how they handle authorization, logging, and cross-tenant isolation. The answer tells you how seriously they take it. Frequently asked questions. What is the Model Context Protocol (MCP)? What did the NSA say about MCP security? What was the Asana MCP data leak? Is MCP safe to use? How can I limit MCP risk? Does MCP affect personal note-taking apps? Should I avoid AI agents because of MCP risks? What this article cites. * Model Context Protocol Security Design GuidanceNSA and CISA · 2026Joint guidance on MCP security risks and recommendations. * The MCP Ecosystem in 2026: State of the StandardChatForest · 2026Industry overview including the Asana incident and adoption figures. * Model Context Protocol SpecificationModel Context Protocol · 2026The July 2026 specification revision and its changes. Related features Built into mindly. Your second brain is one download away. Free for macOS. No account required.

Jobma expands its integration ecosystem to 70+ native integrations, with fraud prevention built in. Asana joins Workday, iCIMS, Greenhouse, and BambooHR in bringing one-way, live, AI interviews and pre-hire assessments into the tools hiring teams already use. With 70+ native integrations, including Workday, Greenhouse, and our newest Asana connection, Jobma brings verified, fraud-resistant interviews right into the tools hiring teams already use." - Mandeep Sodhi MINNETONKA, MN, UNITED STATES, October 30, 2026 /EINPresswire.com/ - Jobma, the intelligent, human-first hiring system, today announced the growth of its integration ecosystem, now spanning more than 70 native integrations across applicant tracking systems (ATS), HRIS, background check providers, job boards, and work management tools. Jobma is the only interviewing and assessment platform that pairs 70+ native integrations with built-in interview fraud prevention, so hiring teams can run verified video interviews and pre-hire assessments without leaving their existing workflow. Most hiring teams already run on an ATS or HRIS, and adding another standalone tool slows them down. Jobma's ecosystem is built to fit into the stack teams already have, with native connections to leading platforms, including: - ATS: Workday, iCIMS, Greenhouse, BambooHR, JazzHR, Bullhorn, Ashby, Oracle Recruiting - HRIS and HR platforms: Deel, Freshteam - Background checks: Checkr, Certn, AccusourceHR - Job boards: LinkedIn, Dice - Work management and collaboration: Asana, Slack, Microsoft Teams - Calendars: Google Calendar, Outlook Calendar Newly added: Asana. Jobma's latest integration brings AI-powered video interviewing and assessments into Asana, one of the world's most popular work management platforms. The integration lets hiring teams send Jobma interview and assessment invites, follow every candidate's status, and jump straight to evaluations without leaving Asana, while leveraging Jobma's identity validation and fraud prevention capabilities throughout the hiring process. Jobma built the integration in-house on Asana's API. Every interview and assessment format, inside the ATS and other HR tools: Jobma's three interview formats connect directly to the systems hiring teams already use, so recruiters can launch interviews and review results without switching platforms: - One-way (pre-recorded) interviews let candidates answer on their own time, so teams can screen high volumes quickly and consistently. - Live interviews bring one-on-one and panel conversations into a structured, recorded format with built-in collaboration. - Autonomous AI interviews use agentic AI to conduct adaptive, structured first-round interviews around the clock, so no applicant waits on a recruiter's calendar. Across every format, Jobma pairs interviews with pre-hire skill assessments and collaborative evaluation. Built-in ID verification and fraud prevention help confirm that the candidate who applies is the candidate who interviews, a growing concern as AI-assisted interview fraud rises. "Talent acquisition teams shouldn't have to rebuild their workflow to evaluate better," said Mandeep Sodhi, President and CEO of Jobma. "With 70+ native integrations, from Workday and Greenhouse to our newest connection with Asana, Jobma meets teams where they already work and brings verified, fraud-resistant interviews right into their process." Jobma's integrations are available now to customers. The full list of partners is at https://www.jobma.com/integrations About Jobma Founded in 2012, Jobma is built for intelligent, human-first hiring. The platform is used by talent acquisition teams at mid-size and large enterprises for pre-recorded, live, and autonomous AI interviews, pre-hire skill assessments, ID verification and fraud prevention, and collaborative candidate evaluation. The platform is SOC 2 Type II and ISO/IEC 27001 certified, compliant with GDPR and CCPA, and aligned with the EU AI Act. Jobma's AI is transparent and human-in-the-loop by design. It never makes hiring decisions automatically. That call always stays with the hiring team. All product and company names are trademarks of their respective owners. Legal Disclaimer: EIN Presswire provides this news content "as is" without warranty of any kind. We do not accept any responsibility or liability for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this article. If you have any complaints or copyright issues related to this article, kindly contact the author above.

Asana schedules December 8 results. Asana stock stands at USD 2.5 billion in value. Published on 10/20/2026 at 22:09 | Editorial responsibility: Rafael Müller, Editor-in-Chief AD HOC NEWS Q2 revenue rose 9.90 percent to USD 216.43 million. Asana stock costs EUR 9.53 on October 20, 2026 at 10:07 p.m. CEST at Lang & Schwarz after hours. Reporting date and operating figures. * Asana's next earnings report is scheduled for December 8, 2026. * Fiscal second-quarter revenue reached USD 216.43 million, up 9.90 percent year over year. * Asana stock costs EUR 9.53 at Lang & Schwarz on October 20, 2026 at 10:07 p.m. CEST. * Adjusted earnings per share were USD 0.10 in fiscal second-quarter 2027. Asana's next earnings report is scheduled for December 8, 2026, and its market capitalization stood at USD 2.5 billion on October 20, 2026, according to Investing.com. Fiscal second-quarter revenue was USD 216.43 million, up 9.90 percent year over year, according to Asana. Discover more Try Team Apps Track Earnings Hire Accountants Asana stock was trading at EUR 9.53 at Lang & Schwarz on October 20, 2026 at 10:07 p.m. CEST in after hours. Revenue growth sets the baseline. The fiscal second-quarter result ended July 31, 2026, gave investors a measurable operating benchmark. Revenue of USD 216.43 million exceeded the USD 214.15 million consensus figure by USD 2.28 million, while adjusted earnings per share reached USD 0.10 against a consensus estimate of USD 0.09, according to MarketBeat. Asana's investor-relations page also lists 100,000 or more customers and a 26.00 percent fiscal second-quarter revenue growth rate. The combination of revenue growth and a positive adjusted earnings figure gives the December report a clear comparison point. NYSE range frames the valuation. On the NYSE, Asana's twelve-month trading range was USD 5.38 to USD 15.28 on October 17, 2026. Market capitalization was USD 2.26 billion on that date, and the price-to-earnings ratio was minus 14.98, according to MarketBeat. These figures are separate from the euro quotation in Germany. What does December 8 add? The scheduled December 8 report will show whether fiscal second-quarter growth carried into the following period. The company previously reported its fiscal second-quarter results on September 3, 2026, and its investor-relations calendar identifies the fiscal year 2027 second-quarter period as ending July 31, 2026. The next report therefore matters for three measurable areas: revenue growth, adjusted earnings per share and customer expansion. The December date is the next concrete corporate milestone currently identified by Investing.com. Discover more Read Economics Stock price and reporting date. For investors in Germany, the euro quotation provides the latest after-hours indication, while the company figures establish the operating reference for the December report. Asana's fiscal second-quarter revenue of USD 216.43 million and adjusted earnings per share of USD 0.10 remain the key published benchmarks. Asana stock facts. * Company: Asana, Inc. * ISIN: US04342Y1047 * Ticker: ASAN * Primary exchange: NYSE * Price Lang & Schwarz (as of October 20, 2026, 10:07 p.m. CEST): EUR 9.53 * Market capitalization: USD 2.5 billion (as of October 20, 2026) * Sector / Industry: Technology / Software - Application * Next earnings date: December 8, 2026 Sponsored Ad Asana stock: new analysis - 22 October. Discover more Invest in Stocks Track Stock Market Try Office Apps Fresh Asana information released. What's the impact for investors? Our latest independent report examines recent figures and market trends. Disclaimer regarding our articles: This is not investment advice, nor is it a recommendation to buy or sell. Information regarding prices, companies, and markets is provided without guarantee; changes may occur at any time. Stock market transactions can result in significant losses. Our articles are created and reviewed, in whole or in part, automatically with the assistance of AI. en | US04342Y1047 | ASAN | boerse | 70378459 | bgmi