Issue Station
Sign In
Tools hereToolbox

gemini-cli - 💡(How to fix) Fix The v0.37.0 release is focused on experimental browser agent, security and user experience enhancements. [1 participants]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
GitHub stats
google-gemini/gemini-cli#25244•Fetched 2026-04-13 05:45:33
View on GitHub
Comments
0
Participants
1
Timeline
17
Reactions
0
Author
xiuwen764-delxiuwen764-del
Participants
xiuwen764-delxiuwen764-del
Timeline (top)
mentioned ×8subscribed ×8labeled ×1
RAW_BUFFERClick to expand / collapse

The v0.37.0 release is focused on experimental browser agent, security and user experience enhancements.

🧪 Experimental

  • 🌐 Browser Agent – Gemini CLI Can now connect and interact with Chrome.
    The experimental Browser Agent is now available on Gemini CLI, and we’d love for you to put it to the test! Gemini CLI can use the Browser Agent to navigate, interact with, and automate complex tasks across the live web—all while keeping your main session context clean and focused. [docs]
  • 🔩 Few things for you to try...
    • Enable the agent by adding "browser_agent": { "enabled": true } to your agents.overrides in settings.json. [docs]
    • Use the @browser_agent command to trigger it for specific tasks, like filling out forms, navigating complex UI, or extracting data.
    • Try natural language requests like "Go to the Gemini CLI release page and summarize the latest changes" to see it in action.

🛡️ Security & File Integrity for Sandboxing – We’re giving you more control over your workflows when using Sandbox mode. [docs]

  • The new forbiddenPaths feature allows the sandbox to automatically respect your project's ignore files (like .gitignore), ensuring the model never touches restricted areas. (#23282, @ehedlund)
  • Additionally, we’ve introduced a dedicated lockdown for environment files to keep your secrets and API keys strictly private. (#23712, @DavidAPierce)

✨ Refreshed UI & Navigation

  • Tokyo Night Theme: A beautiful new dark theme option is now available. (#24054, @danrneal)
  • Accelerated Scrolling: Improved navigation with accelerated scrolling performance in alternate buffer modes. (#23940, @devr0306 )
  • Clearer Status: The footer now explicitly displays your current Authentication status and provides more compact tool output layouts. (#24042, @sehoon38 )
  • Immediate Queueing: You can now use the 'Tab' key to queue up messages while the model is still generating a response. (#23661, @sehoon38)

🧭 Platform Support

  • Windows & Linux Expansion: Added support for dynamic sandbox expansion and Git Worktree support for both Linux and Windows. (#23692, @galz10 )
  • Model Routing: Improved dynamic routing for Gemini 3.1 Pro and enabled mid-stream retries for better resilience. (#23641, @kevinjwang1 )

🎉 Gemini CLI Extensions

  • Google Cloud (gcloud): Enable MCP-compatible AI agents to interact with your Google Cloud environment using natural language. Automate complex workflows and manage cloud resources without memorizing complex command syntax.
    gemini extensions install https://github.com/gemini-cli-extensions/gcloud

Small features, Polish, Reliability & Bug Fixes: A large amount of changes, smaller features, UI updates, reliability and bug fixes + general polish made it in this week!

Originally posted by @srithreepo in https://github.com/google-gemini/gemini-cli/discussions/25064

extent analysis

TL;DR

To utilize the new features in the v0.37.0 release, such as the experimental Browser Agent, users should enable the agent by modifying their settings.json file and explore the new functionalities through the provided documentation and commands.

Guidance

  • To start using the Browser Agent, add "browser_agent": { "enabled": true } to your agents.overrides in settings.json as indicated in the documentation.
  • Explore the @browser_agent command for triggering specific tasks like form filling, navigation, or data extraction.
  • Review the documentation for Security & File Integrity enhancements, especially the forbiddenPaths feature and lockdown for environment files, to understand how to better control your workflows in Sandbox mode.
  • Test the new UI features, including the Tokyo Night Theme, accelerated scrolling, and clearer status displays, to improve your overall user experience.

Example

No specific code snippet is provided due to the nature of the issue focusing on release notes and feature updates rather than a specific coding problem.

Notes

The guidance provided is based on the information given in the release notes. Users should refer to the official documentation and GitHub discussions for more detailed instructions and potential updates.

Recommendation

Apply the workaround by enabling the Browser Agent and exploring its features as described, to take advantage of the new functionalities and improvements in the v0.37.0 release. This allows users to leverage the enhanced security, user experience, and experimental features without waiting for further updates.

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…
#api#ssr#permission error#memory optimization#batch processing

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING