Loading story…
1Password and Anthropic launched 1Password for Claude, a browser integration that lets Claude access accounts protected by credentials stored in 1Password without exposing passwords or MFA codes to the AI model. The system requires the agent to request access for a specific task, after which the user must approve the request—reportedly including biometric confirmation in some flows—and 1Password injects the credentials directly into the destination site through a secure channel outside the model’s view.
The companies said the feature uses a per-session, task-scoped authorization model with no standing access, forcing Claude to request approval again for later attempts. 1Password also introduced a broader Agentic Mode that automatically locks down vault access when a recognized AI agent controls the browser, limiting the agent to only the explicitly approved credentials for the current task. The Claude integration is launching first, with availability on Mac for 1Password business, family, and individual plans, and broader support for other browser-based agents planned later.
Mallory correlates global threat intelligence with your attack surface — know if you’re exposed before adversaries strike.
1 event from the most recent confirmed update back to the earliest known activity.
1Password and Anthropic announced 1Password for Claude, a browser integration that lets Claude use credentials stored in 1Password vaults without exposing passwords or MFA codes to the model. The launch also introduced Agentic Mode, which restricts browser credential access to user-approved, task-scoped sessions when an AI agent is controlling the browser.
Vulnerabilities, threat actors, malware, products, organizations, and breaches Mallory has linked to this story.
6 references tracked. Mallory keeps watching after this page renders.
thenewstack.io
Open sourcehelpnetsecurity.com
Open sourceitpro.com
Open sourcezdnet.com
Open sourcemacrumors.com
Open source1password.com
Open sourceMap indicators from this story to your assets and identify affected systems in minutes.
Every observed campaign, victim, and pivot linked to actors named in this story.
Malware, exploits, and IOCs connected to the activity described here.
YARA, Sigma, and Snort rules deployed to your SIEM as soon as they’re published.
Get matching new stories delivered to your team as they break — not the next morning.
Ask questions about this story and take action on the answers.
Other stories Mallory is tracking that overlap on entities, sources, or topic labels.
1Password has introduced a new security feature called Secure Agentic Autofill, designed to address the growing risks associated with AI-powered browser agents in enterprise environments. As organizations increasingly rely on agentic browsers—browsers that use artificial intelligence agents to automate tasks—new authentication challenges have emerged, particularly the risk of credential leakage. AI browser agents, while efficient at automating repetitive tasks such as form filling, appointment scheduling, and online purchases, often store sensitive credentials and secrets in browser session storage or cookies, which are frequently unencrypted and vulnerable to attack. Attackers have begun to exploit these weaknesses, recognizing that more enterprise work is conducted within browsers, thereby increasing the attack surface. To mitigate these risks, 1Password has partnered with Browserbase to launch Secure Agentic Autofill, a capability that allows AI agents to authenticate in browsers without exposing user credentials. This feature incorporates a human-in-the-loop approach, ensuring that sensitive authentication processes remain under user control and oversight. The solution aims to provide enterprises with greater visibility, control, and governance over how AI agents access internal systems and sensitive data. Nancy Wang, SVP of engineering at 1Password, emphasized that Secure Agentic Autofill is designed to solve the pressing issues of credential exposure and unauthorized access by keeping humans involved in the authentication loop. The feature is particularly relevant as a new generation of agentic browsers, such as Perplexity’s Comet and Opera Neon, gain popularity and are integrated into enterprise workflows. By preventing AI agents from directly accessing or storing credentials, 1Password’s solution reduces the risk of credential theft and subsequent breaches. The integration with Browserbase further enhances the security posture by ensuring that only authorized agents, under human supervision, can perform sensitive operations. This development reflects a broader industry trend of addressing the security implications of AI-driven automation in the workplace. Enterprises adopting AI browser agents are encouraged to implement solutions like Secure Agentic Autofill to maintain robust authentication security. The launch of this feature demonstrates 1Password’s commitment to proactively addressing emerging threats in the evolving landscape of identity and access management. As AI agents become more prevalent, the need for secure, user-controlled authentication mechanisms will continue to grow. 1Password’s Secure Agentic Autofill sets a new standard for protecting credentials in environments where AI automation is increasingly common. The feature is expected to help organizations balance the benefits of AI-driven productivity with the imperative of maintaining strong security controls. By keeping a human in the loop, 1Password ensures that enterprises can safely leverage AI agents without compromising sensitive information.
Jun 29, 2026
Anthropic announced **Claude Code Security**, an embedded capability in *Claude Code* that scans customer codebases for vulnerabilities and suggests patches, initially rolling out to a limited set of enterprise/team customers for testing. The company said the feature was stress-tested via internal red-teaming, Capture-the-Flag exercises, and collaboration with **Pacific Northwest National Laboratory**, and positioned it as a way to reduce reliance on manual security reviews as AI-assisted “vibe coding” increases and attackers also use AI to accelerate weakness discovery. In parallel, Anthropic released **Claude Sonnet 4.6**, emphasizing improved coding performance, stronger “computer use” capabilities, and expanded developer tooling (e.g., adaptive/extended thinking modes, beta context compaction, and API tools for web search/fetch and code execution). Separate commentary highlighted the security risk of **agentic coding assistants** (e.g., *Claude Code*, *Cursor*, *GitHub Copilot*) operating with broad privileges—file access, shell execution, and secret handling—and argued that the emerging **Model Context Protocol (MCP)** ecosystem needs stronger, future-proof identity controls; additional industry guidance promoted **MLSecOps** as a way to integrate security into AI/ML development lifecycles, though it did not report a specific incident or vulnerability.
Jun 29, 2026
Anthropic released an update that lets **Claude** interact with third-party workplace applications *inside* the chat interface using the open **Model Context Protocol (MCP)**, reducing the need to switch between separate app tabs. The capability supports interactive embedded experiences (e.g., project boards, dashboards, design canvases, messaging) so users can view and edit content from services such as **Slack, Asana, Box, Canva, Figma, Hex, and monday.com** directly within Claude, including workflows like drafting incident communications or updating task timelines. The implementation is powered by an MCP extension described as **MCP Apps**, which allows an MCP server to provide an interactive UI that can render and accept user actions within an AI product. ZDNET characterized the move as part of Anthropic’s broader push to make Claude a primary work interface, building on earlier “connector” approaches that primarily moved data via prompts; the new approach enables more direct, app-native style interaction within the Claude conversation. A separate ZDNET piece offering general “IT playbook” guidance for the AI era does not provide substantive details about this specific Claude/MCP integration update.
Jun 29, 2026