This release marks a major step forward in making Cortex feel more like a dependable engineering and security partner inside the developer environment. The update focuses on three outcomes: better understanding of the user’s project, smoother handling of code changes, and a more controlled experience when the assistant takes action.
🧠 Smarter Project Understanding and Context Awareness
Cortex now has a stronger foundation for understanding project context. Instead of relying only on the visible file or manually attached snippets, it can use structured workspace signals to decide what information is relevant to a request. This helps the assistant respond more accurately to questions about bugs, implementation work, architecture, dependencies, and security posture.
🔐 AI Threat Model and AI Security Review
The release also introduces two new security workflows: AI Threat Model and AI Security Review. AI Threat Model helps teams reason about how a system could be attacked before issues become incidents, including sensitive assets, trust boundaries, exposed interfaces, data movement, and likely abuse paths. AI Security Review focuses on implementation risk, helping developers inspect code for common vulnerability classes, unsafe assumptions, weak controls, and security gaps that may not be obvious during normal development.
AI Threat Model is designed to help teams shift security thinking earlier in the development cycle. Instead of waiting for a vulnerability scan or late-stage review, developers can ask Cortex to reason about how an application, feature, API, or integration could be misused. It can help identify high-value assets, external entry points, trust boundaries, privileged operations, sensitive data flows, and attacker goals. This makes threat modeling more accessible during everyday engineering work, especially when teams are changing authentication flows, adding new services, exposing APIs, or introducing third-party integrations.
AI Security Review is focused on code-level and implementation-level risk. It helps developers review concrete changes or files for security weaknesses such as missing authorization checks, unsafe input handling, injection risks, insecure defaults, secret exposure, weak validation, risky dependency usage, and incomplete error handling. The goal is not to replace formal security review, but to give developers a fast security-aware pass while they are already in the editor, so potential issues can be caught and fixed before they become larger review, testing, or production problems.
💻 More Natural Coding Workflows
The coding workflow is also more natural. When users ask for changes, Cortex can now handle edits in a way that feels closer to how developers already work: changes can be applied, reviewed, undone, redone, and inspected through diffs. The experience is designed to reduce unnecessary confirmation steps while still preserving user control when it matters.
🛡️ Safer Local Action Handling
A key part of this release is safer action handling. When Cortex needs to perform a local operation, the extension now manages that action through the VS Code side of the experience. Users can approve individual actions or allow a category of action for the current session, while permissions remain temporary and local rather than permanently stored.
🔎 Improved Privacy and Context Protection
Privacy and context handling were also improved. The extension has a clearer separation between general chat behavior, privacy scanning, and tool execution. This makes it easier to protect sensitive content while still giving the assistant enough information to be useful on real projects, including repositories that contain credentials, security findings, or sensitive implementation details.
⚡ Refined Chat and Tooling Experience
The chat experience has been polished across streaming, tool activity, status messages, retry behavior, cancellation, reloads, and diff display. Many of these changes are small individually, but together they make the assistant feel less fragile and more predictable during longer coding sessions.
🌐 Moving Toward Production Ready Agentic Security Workflows
Overall, this update moves Cortex closer to a production-ready agentic coding and security workflow. It improves context quality, makes code edits easier to trust, reduces noisy UI behavior, and sets up a cleaner foundation for future capabilities such as richer file operations, git-aware workflows, deeper security analysis, and higher-level implementation plans.
