This is the biggest update to Intelligent Terminal since we launched. This release introduces the new sidebar for managing your tabs and agents, including recent agent sessions and agents running in other terminal windows. You also get rich tab information you can customize, global search, and a filter for focusing on your agents. Intelligent Terminal now detects installed agent CLIs and adds them to the new tab menu, while headless mode keeps selected tabs running after you close the terminal.

You can grab the update from the Microsoft Store, via winget install Microsoft.IntelligentTerminal, or by heading to the release page on GitHub. Let’s dive in!

Introducing the sidebar

We wanted a richer, easier-to-read experience for working with agents, and we heard that the classic tab bar needed more room for the context agent workflows bring. The new sidebar expands your tabs into a vertical view where agents and shells are easier to scan, manage, and return to. Every tab has room for more than a truncated title, with agent CLI sessions and terminals together in the same list.

These are still your tabs, so everything works the way it already does: click to select, × to close, drag to reorder, right-click for the tab menu.

Here’s what it brings that’s new:

• Live agent status on every tab, so you can see what an agent is up to without opening it, with an attention dot on any tab waiting on you
• Pinning by right-clicking a tab to pin the tab you’re watching at the top of the sidebar
• Keep a tab running in headless mode by right-clicking a tab > headless mode even after you close the window
• Split panes nested under their tab, so you always have visibility into what each pane is doing, and jumping to it directly
• Global search across your open tabs and your recent agent sessions at once
• An agent filter to cut the list down to just your agent sessions
• Recent agent sessions in a collapsible section under your open tabs, including agents running in other terminal windows

The sidebar is on by default. If you prefer the classic horizontal tabs, right-click any tab and select Switch to horizontal tabs, or change the layout under Settings > Appearance > Tab Layout. You can switch back to the sidebar anytime, resize it, or collapse it with the sidebar button or Ctrl+Shift+S.

Acrylic theming is also on by default for the sidebar. If you prefer a solid background, turn off Use acrylic material in the tab row under Settings > Appearance.

Rich tabs

Tabs show status and their current working directory by default. When a tab is running an agent CLI, the shell icon swaps for the agent’s icon and the status reads the latest thing the agent is doing, so you can tell whether it’s working, waiting on your input, or asking to run a tool without opening it.

You can swap either field out for something you’d rather see by clicking the filter icon in the tab header:

• Agent status to see what the agent is doing
• Current working directory to see where the tab is working
• Git information such as repository, branch, or changes to have glanceable repository information

Pick up to two fields from the filter menu in the sidebar. Turning one on swaps another out so your tabs stay readable.

Search your tabs and your past sessions

Type to narrow the list as you go. Search is global, so it looks through the tabs you have open in your window and the agent sessions you have open across all windows and sessions you’ve already closed, at the same time. Whether the thing you’re after is running right now or finished yesterday, it shows up in a single quick search.

Filter down to just your agents

Open the filter menu and turn on Agents only to cut the list down to the tabs running an agent CLI. Turn it off to bring all your terminal tabs back.

Find recent agents across your terminal windows

Open the filter menu and turn on Recent agent sessions to add a collapsible list under your open tabs in the sidebar. It shows prior agent sessions and agents running in other terminal windows, with their title, working directory, and relative timestamp. Select one to reopen recent work in a new tab or jump focus straight to an agent already open and running elsewhere.

Start your agent from the new tab menu

Intelligent Terminal already detects supported agent CLIs installed on your machine and now it creates profiles for them automatically. Open the new tab menu and start GitHub Copilot CLI, Claude, or another detected agent the same way you start PowerShell or any other terminal profile.

Keep work running in headless mode

A long agent task or a long build used to die the moment you closed the terminal. Headless mode lets it continue in the background.

Right-click a tab and select Turn on headless mode. You can close the terminal and walk away while the session keeps running in the background until the machine stops.

Easy to find: tabs running in headless mode show up in the sidebar will running and the system tray when there are no active windows. Agent sessions (agent CLIs running in a tab) that are in headless mode will always show up under your Recent agent sessions in the sidebar when they are closed.

A huge thanks to our community

Thank you for constantly providing feedback and helping shape Intelligent Terminal. For everything else in this release, see the full release notes on GitHub.

To see Intelligent Terminal in action entirely with local models, check out Running a 120B model offline on Surface RTX Spark Dev Box with Intelligent Terminal:

We’ll keep iterating rapidly on bugs and feature requests while Intelligent Terminal is in experimental. If you have a feature request or find a bug, submit an issue or visit our GitHub repository. Thank you!

The post Intelligent Terminal 0.3: A new way to manage your tabs and agents appeared first on Windows Command Line.

Dr Andrew Tridgell of the Ardupilot project discusses flight control software for autonomous and remote controlled aircraft. This field has advanced considerably in recent years, and autonomous vehicles are playing an increasingly important role in many industries -- from package delivery to scientific surveying and film-making -- as well as in the consumer market.

Host Jason Davey speaks with Tridgell about the fundamental concepts required to keep a drone in the air, the hardware and software architecture behind these vehicles, and how development and testing are handled in Ardupilot, one of the leading open source flight control software stacks.

Download audio: https://traffic.libsyn.com/secure/seradio/745-andrew-tridgell-flight-control-software.mp3?dest-id=23379

Microsoft's business unit shuffle and shrouded AI spending spark a sharp conversation on whether shareholders and users can trust what's hidden behind new financial reports. Also, Start11 v3 adds Start widgets and menu sharing, plus Taskbar Islands, app groups, and more. Finally, LM Studio is the obvious place to start with local AI, and it works on any PC (not just an RTX Spark). Just be sure to get the right app.

Microsoft Earnings

• Microsoft earned a net income of 36.6 billion on revenues of $92.3 billion in the quarter ending September 30, 2026
• Those are both double-digit gains over the year-ago quarter, or 32 percent and 19 percent growth, respectively, year-over-year (YOY)
• We were promised more transparency, but what we got is a mixed bag
• Microsoft is too complicated a business for only two top-level business segments
• Windows revenues from PC makers down 27 percent (!)
• XBOX gets a hard revenue number, and it's not bad
• Microsoft spent $45.2 billion on capex (AI) expenses, up 30 percent - this is not the real number - and it will get bigger
• No paid Copilot numbers, weird
• Microsoft "earned" $10 billion in revenue from OpenAI - OpenAI still owes Microsoft money - MS stake in OpenAI is down; no one said why

Windows

• We talked about this last week, but I now have two sources at Microsoft confirming that Microsoft's AI strategy is the reason it's fixing Windows 11
• Microsoft discusses its work to make Windows 11 fit in 8 GB of RAM
• Microsoft Edge is the first Windows browser to support accelerated app launches
• Experimental and Experimental 26H1 get new inline actions for the new Windows Search

Hardware

• It's been weird watching other reviewers try to figure out RTX Spark, but this much is clear: This platform is not ready
• Baby steps: A few minor victories after weeks of defeats

More earnings

• AMD up 45 percent to $13.4 billion
• Intel up 21 percent to $16.5 billion
• Google/Alphabet up 25 percent to $127.6 billion
• Apple up 12 percent to $114.5 billion
• Amazon up 12 percent to $202 billion

AI/dev

• Three and a half years after it should have happened, Anthropic, Google, Microsoft, and OpenAI agree to let sites opt out of scraping
• OpenAI to unify chat and Work but not Codex
• Google says schools that disable AI will lose access to Google Search - They're integrated. It's nice to see Google channeling Microsoft again
• .NET Conf is next week, and Microsoft just posted the schedule

XBOX and gaming

• There was a bit of drama when XBOX COO left after only a few months on the job, and now there's more
• New titles coming to Game Pass in early November
• The next Elder Scrolls game is ES VI Sentinel
• GTA VI to rival Call of Duty - in the amount of disk space it requires

Tips and picks

• App pick of the week: Start11 v3
• App pick of the week #2: LM Studio
• RunAs Radio this week: PowerShell 7.7 and DSC 3.3 with Jason Helmick
• Brown liquor pick of the week: Caol Ila 12

*Thumbnail courtesy of Joe Esposito in Club TWiT

These show notes have been truncated due to length. For the full show notes, visit https://twit.tv/shows/windows-weekly/episodes/1008

Hosts: Paul Thurrott, Richard Campbell, and Mikah Sargent

Sponsors:

• helixsleep.com/windows
• originhq.com/windowsweekly
• trustedtech.team/twit
• canary.tools/twit - use code: TWIT

Download audio: https://pdst.fm/e/pscrb.fm/rss/p/mgln.ai/e/294/cdn.twit.tv/cap/ww_1008/4db0cdf9-96ab-42c6-8d11-3613238d0151.mp3

Coding agents write code faster than ever, and the bottleneck in shipping software has moved to validating it. A 15-minute build gives an agent 4 feedback cycles an hour. A 5-minute build gives it 12. Romain sits down with Dan Manges, co-founder and CEO of RWX and former CTO of Braintree and Root, to talk about what CI/CD looks like when agents are the main users, and how the rest of the software development lifecycle has to change with it. Key topics covered: - Why traditional CI/CD pipelines slow down coding agents - Graph-based execution and content-based caching, compared to Docker's lineage-based caching - A CLI-first workflow: run CI against local changes without committing or pushing, and get a green PR check right away if the agent already ran it - Sandboxes for the agent's inner loop, MCP, semantic outputs, and retrying only the failed tests - Building RWX on AWS: a distributed container runtime with no image build and push, a hot pool of EC2 M8a instances, single-tenant VMs, and self-hosted runners in customer VPCs - Rethinking the SDLC from first principles: give agents the context to get it right, throw code away and fix AGENTS.md instead of iterating - Software factories for internal tools versus human review for mission-critical code - Turning large changes into reviewable work by building the whole change first, then extracting shippable pieces - Guardrails for agent-driven changes: CloudWatch, gradual rollouts and feature flags - Results: pipelines about 2x faster on average, and about 35% lower cost - Is infrastructure as code still necessary? Chapters: 00:00 Rethinking engineering for AI agents 00:52 Meet Dan Manges and his engineering journey 02:59 Why traditional CI/CD becomes a bottleneck 05:34 Faster feedback loops for coding agents 08:02 Graph execution and content-based caching 13:19 A CLI-first, agent-native CI workflow 15:33 Unifying agent development and CI environments 17:45 Building RWX on AWS 21:33 Rethinking the SDLC for agents 24:33 When humans should review agent code 27:07 Giving agents the context to get it right 29:29 Sharing agent skills and team practices 35:50 Building software factories with agents 40:32 Turning large changes into reviewable work 46:20 Guardrails for safe agent-driven changes 48:06 Measuring CI speed, cost, and agent benefits 51:10 Signals your CI pipeline needs rethinking 54:02 Is infrastructure as code still necessary? 56:38 Closing thoughts and recommendations

With Dan Manges, Co-founder and CEO of RWX

RWX

RWX documentation

RWX on AWS Marketplace

Work in progress commits are a thing of the past (RWX blog)

Amazon EC2 M8a instances

Amazon CloudWatch

Bazel

Nix

Amp

Sapiens: A Brief History of Humankind, Yuval Noah Harari

Download audio: https://op3.dev/e/dts.podtrac.com/redirect.mp3/developers.podcast.go-aws.com/media/227.mp3

Rewriting is where good stories become better. Use this ultimate rewriting checklist for authors to polish your prose.

The first draft of your novel is a sketch. It is a rough piece of writing where you have worked out a shape of a story. Once you have completed this, the real work begins.

Richard North Patterson says: ‘Writing is rewriting. A writer must learn to deepen characters, trim writing, intensify scenes. To fall in love with the first draft to the point where one cannot change it is to greatly enhance the prospects of never publishing.’

Many authors rewrite their manuscripts many times before they are happy to send their book to the publisher.

Michael Crichton said: ‘Books aren’t written – they’re rewritten. Including your own. It is one of the hardest things to accept, especially after the seventh rewrite hasn’t quite done it.’

Rewriting is an essential part of the writing process. Use this checklist to help you identify weaknesses, tighten your prose, and make sure your story delivers on its promise. Tick the boxes as you go along. (Download your own list at the end of this checklist.)

The Ultimate Rewriting Checklist For Authors

Story And Structure

1. What is your promise to the reader? Is it clear from the beginning? ☐
2. Does your story begin at the right moment? ☐
3. Have you checked the requirements of your genre? ☐
4. Have you introduced your main characters in the opening scenes? ☐
5. Have you given your protagonist a distinctive voice? ☐
6. Are your protagonist and antagonist believable? ☐
7. Is there enough conflict? ☐
8. Do you have a plot? ☐
9. Are there holes in your plot? ☐

Viewpoint, Setting, And Dialogue

10. Have you chosen the correct viewpoint for your novel? ☐
11. Have you placed your characters firmly in time and place? ☐
12. Is there enough dialogue for your story and genre? ☐
13. Does your dialogue flow? Have you read it aloud? ☐
14. Have you removed information repeated in dialogue and narration? ☐
15. Does your use of contractions sound natural for your characters and narrative voice? ☐

Editing Your Prose

16. Have you revised your novel as many times as needed? ☐
17. Do you have too many long sentences? ☐
18. Have you removed unnecessary adverbs? ☐
19. Have you removed unnecessary adjectives? ☐
20. Have you replaced vague, confusing, or illogical writing with clear, precise language? ☐
21. Have you cut out gratuitous similes and metaphors? ☐
22. Does your use of profanity suit your characters and story? ☐
23. Have you used the passive voice deliberately and effectively? ☐
24. Have you removed unnecessarily elaborate writing? ☐
25. Have you removed author intrusion and inadvertent preaching? ☐
26. Have you used the correct lexicon for your genre? ☐
27. Have you removed unnecessary modifiers and qualifiers? ☐
28. Have you done a spell check? ☐
29. Have you read your manuscript aloud to catch awkward phrasing, repetition, and inconsistencies? ☐
30. Have you made good on your promise? Is your reader satisfied? ☐

Download your checklist here: Rewriting Checklist For Authors

The Last Word

Rewriting takes time, patience, and a willingness to be ruthless with your words. Use this checklist to work through your manuscript, strengthen your story, and polish your prose. Remember, every rewrite brings you closer to the book you want to write.

You may enjoy these useful posts:

1. An Editing Checklist For Writers
2. 12-Steps to Self-Editing – your stress-free guide to preparing a manuscript
3. What 20 Famous Authors Had to Say About Rewriting

by Amanda Patterson
© Amanda Patterson

More posts from Amanda:

1. 14 Popular Crime Fiction Sub-Genres
2. How To Use Secrets To Create Suspense
3. 10 Ways To Start A Horror Story
4. What’s The Difference Between A Commercial And A Literary Plot?
5. 7 Secrets To Writing Happy Characters Without Boring Your Audience
6. What Is A Phobia? How To Write Characters With Phobias
7. 3 Simple Ways To Make Your Hero Fight Back
8. 7 Reasons To Write The Ending Of Your Novel First
9. How To Make Readers Care About An Amoral Protagonist
10. Why Revenge Is A Brilliant Plot For First-Time Writers

Top Tip: Sign up for our free daily writing links.

The post The Ultimate Rewriting Checklist For Authors appeared first on Writers Write.

There’s a good amount of advice in today’s reading list. Now that we’re deeper into the mainstreaming of AI, the advice is feeling more grounded in real experience.

[blog] AI Can Write More of the Code. We Still Have to Engineer the System. Reasonable point of view. We’ve got eons more automation with AI reasoning mixed in, but human engineers want to define the work and be able to verify the results.

[blog] Five rules for evaluating agent skills before they bloat your context window. Terrific post with genuinely useful (and proven) approaches for evaluating skills to see if they’re worth using.

[article] Influencing the agents. Great post from Matt. The agents are making default decisions on products, and many of us will just accept those recommendations for convenience. Build and explain your products accordingly.

[blog] Watch Henry Cavill build a video game on his Googlebook. Big fan of his, and glad he’s doing fun work with us.

[blog] ADK2.0 101 (#4a): Evaluating your agents. Even if you’re not building them yourself (yet), get comfortable with the ideas behind agent evals.

[blog] How AI Applications Earn Their Premium. Where does value live now? Why buy versus build it yourself? These seem like the right levers.

[blog] Guest Posts: How //go:embed Works: From Source Files to Binary. Other programming languages let you stash content in the resulting binary, but Go’s approach is simple, powerful, and somewhat unique.

[blog] Top Gen AI Frameworks for Go in 2026: A Hands-On Comparison. I don’t know what’s going to happen with agent frameworks. We probably have too many. Even too few? It’s possible that general agents, given skills, will negate the need for many custom-coded agents. Or not.

[blog] Vibe Coding, Meet Reality. It’s about layered infrastructure. Not having people vibe a whole stack. Are we giving people the right places to build, and supporting them with safe and effective substrates?

[blog] AlloyDB Express: Instant, scale-to-zero PostgreSQL for developers and coding agents. Many services offer fast database instances now. We’re a bit unique in offering built-in AI capabilities and one-click upgrade to production-scale. Comes with a free tier too.

[article] Jensen Huang’s Advice to Young Developers: “Become the World’s Best User of AI.” Optimistic take, but this is similar advice I give when asked by young people. You are capable of more than ever.

Want to get this update sent to you every day? Subscribe to my RSS feed or subscribe via email below: