You can now review the code written by Claude Code, Codex, OpenCode, and more directly in Warp, and send inline comments straight to the agent.
It's like a PR review without having to leave your terminal. Try it out!
You can now attach past conversations as context with `@ conversation` using Warp's agent.
Very helpful after compaction, or when splitting new conversations to reset context.
You can now pull GitHub comments directly into your agent conversations with /pr-comments.
✅ View comments in-line with Warp's code review pane
✅ Ask the agent to weigh in and address feedback
Now in preview: see the PR your agent is working on straight from your terminal input.
Live for the Warp agent, with other coding agents coming next.
You can shift-enter in the OpenCode input box now 🤝
Long version: We now support the kitty keyboard protocol. This fixes a number of keyboard input issues for interactive CLIs, including your favorite coding agents.
Computer use is a huge deal. It lets agents close the loop by clicking around apps they build, verify changes e2e, and screenshot changes for review.
Here's a technical deep dive from Daniel Peng (Warp eng) of how we built model-agnostic computer use for cloud agents 🧵
Warp now has LSP support across the code review panel and file viewer!
✅ Support for JS, Python, Go, Rust, and more to come
✅ Find references, go-to definition, hover hints for types, and compile-on-save
And, bonus customization: you can now hide the "code review" button if you prefer. Go to Settings > Appearance > Tabs.
It'll always be accessible with the ⌘⇧+ shortcut
You can now customize the chips you see for the built-in terminal prompt!
✅ Add or remove chips that are most useful to you
✅ Rearrange chips for language version, directory, git branch, and more
Just go to Settings > Appearance in the latest version
oh snap,
JIT MCP
Warp doesn't make me install MCP servers, it offers to install them, just in time, when it needs the tool. Great UX.
nerdy.dev/try-warp
We're working on table rendering! Stay tuned
We're giving free credits to high-impact open source projects using Oz cloud agents.
@tanstack.com, @vite.dev, PicGo, and other major projects are already prototyping with Oz to automate issue triage, PR reviews, support, and more.
Interested maintainer? Tell us about your project 👇
We open sourced the Skills we use for our coding agents at Warp.
These cover the most useful automations, like auditing accessibility, updating docs, improving test coverage, and more.
Install into your favorite coding agent, or reference them in your next Oz agent run
github.com/warpdotdev/o...
We’re giving 1,000 bonus credits for cloud agents to everyone who upgrades to Build or Max in February.
Get started at oz.dev and tell us what you’re building!
Oz also helps you turn Skills into repeatable automations.
Set up cron schedules to analyze logs, clean up stale feature flags, check for vulnerabilities, or anything else you want to automate.
With Oz, any Skill can be launched as an agent.
Use our prebuilt Skills for:
✅ issue triage
✅ accessibility audits
✅ test generation
Or, commit the Skills that you already use with your coding agents.
Oz doesn’t limit your environments to a single git repository.
With Docker sandboxing, agents can work across your entire suite of repos and open multiple PRs from a single prompt.
Plus, the Dockerfile is all yours. Ask your agent to add dependencies using the oz CLI.
Use the management UI to review every cloud agent run across your team.
You get a live session link for every run so you can interact with the agent as it works. When it’s done, you can review artifacts including plans and open pull requests.
In Warp, the web, or your phone!
Oz is fully programmable with the `oz` CLI.
✅ `oz agent run-cloud` to kick off a cloud agent with a prompt and optional Skill
✅ `oz schedule create` to set up automated tasks
✅ `oz environment create / update` to manage cloud sandboxes
Introducing Oz: the platform to orchestrate agents in the cloud.
Spin up hundreds of agents from your terminal, browser, the API, or your phone. Each agent gets a Docker environment to build, test, and write PRs.
Here's the blog post to learn more about how it works 👇
warp.dev/blog/oz-orch...
Blue sky background with a cloud, a cloud in the shape of a mouse, and a cloud in the shape of an... armadillo? Date says 02.10.26
Launching something big tomorrow...
This is tracked!
I’m excited to share that I’ve joined the team at @warp.dev.
I’ve long loved the terminal and I appreciate the power of AI-enabled software development. Warp brings those together in a way I’m excited to learn from and help build with a super product-focused team.
💻️ 🐚 ❤️ 🤖 🚀
Introducing Agents 3.0: Four new features expand what agents can do and how you can interact with them.
- Full Terminal Use to use REPLs, debuggers, and more
- /plan for spec-driven development
- Interactive Code Review
- New Slack, Linear, and GitHub Actions integrations
Warp’s CLI is still in beta— join the #feedback-warp-cli channel in the Warp community Slack to provide feedback, get help, and let us know what features you want to see next:
go.warp.dev/join-preview
Introducing Warp CLI: a new way to use Warp’s agents.
You can now access Warp’s agents from any environment.
Use the CLI to build custom integrations, use Warp on remote machines, trigger it when CI fails, or build your own Slack integrations.
youtu.be/hV6UdEf3C1I
New Youtube video up on how to make Warp's UI more minimal.
Since we've gotten user feedback about how Warp can be overwhelming/cluttered at times 👇
youtu.be/1GKsIT8FSsE?...
