Devin upgraded its security agent with deeper scanning and autonomous remediation, while GitHub made Copilot assisted permissions generally available.
AI coding tools5 items
OpenAI also shared examples of coding agents supporting incident investigation and security analysis.
Devin Security 2.0 adds frontier-model scanning and autonomous remediation
Cognition announced a major Devin Security upgrade combining frontier cyber-model scans, environment-aware prioritization, and autonomous remediation that turns findings into pull requests. It also announced hundreds of millions of dollars in Devin credits to help enterprises find and fix critical vulnerabilities.
Why it matters: The upgrade links vulnerability discovery to proposed code changes, potentially reducing the manual work between security findings and remediation.
GitHub Copilot assisted permissions reach general availability across CLI, app and VS Code
Assisted permissions are now generally available in GitHub Copilot CLI, the Copilot app, and Visual Studio Code. The feature applies when Copilot requests permission for actions such as running commands or accessing file-system paths.
Why it matters: Teams using these Copilot interfaces can now incorporate the generally available feature into their agent permission workflows.
OpenAI reports Codex Ultrafast cuts incident investigation from hours to minutes
OpenAI described engineers using Codex Ultrafast to investigate incidents in minutes rather than hours. The account also covers security analysis, code refactoring, and daily engineering workflows.
Why it matters: The account offers a concrete operational use case for coding agents beyond writing code, though the reported gains come from OpenAI itself.
Greptile uses GPT-6 Astra to test workflows beyond code review
OpenAI described Greptile using GPT-6 Astra to test user workflows, Daybreak to strengthen security, and Codex to reduce rework.
Why it matters: Workflow testing broadens agent-assisted bug detection beyond reviewing code changes alone.
CodeRabbit uses GPT-6.1 Sol for security investigations across complex codebases
CodeRabbit uses GPT-6.1 Sol to investigate security risks across complex codebases. Its security agent helps developers review findings and proposed fixes.
Why it matters: The example highlights a security workflow that keeps developers involved in reviewing agent findings and fixes.