Claude Cowork earns its place: AI tool tackles email chaos and saves hours on tedious tasks

I gave Claude Cowork 7 non-coding jobs, and it earned a spot in my toolbox

Follow ZDNET: Add us as a preferred source on Google. ZDNET's key takeaways * Claude Cowork turned messy tasks into useful results. * It helped fix servers, sort PDFs, and review contracts. * Giving an AI that much access remains unsettling. I am not a naturally trusting person. So when Claude Cowork was launched, the idea of giving an AI access to my Google Docs and Gmail did not sit well with me. Heck, back in the day, the idea of letting Google manage my email didn't sit well with me, either. I was once a big proponent of only using servers you could touch, restart, and disassemble at 3 a.m. The idea of letting Google, the self-acknowledged master of information vacuuming, have access to my email seemed ludicrous. Heck, even letting a cloud hosting provider host my web server seemed extreme. There was a time in the early days of the web when my startup's $2,000-a-month T-1 line entered my apartment in my bedroom, ran through my bathroom, then through my bedroom closet, across the hall, and into a linen closet with an octet of tower servers and an ambient temperature in the mid-90s. Try explaining that to your apartment manager. And no, it wasn't a grow room, which I had to prove to multiple authorities over the years. My only saving grace was I could fix their computers for them. Also: I tested the new Claude Desktop on Linux - here's how it compares to rival apps The point is that I'm a bit of a control freak. Letting Claude Cowork loose to do work for me triggered my control freak alarms in a big way. But it's been slowly worming its way into my life the same way cloud hosting providers and Gmail did: it's able to save me time. I'm still not comfortable with it, but over the past few months, I've set it loose on some projects where it came through, saving me hours of tedious work. In this article, I'm going to give you a rapid-fire summary of seven of those projects. If you want more information, I've already written detailed articles about a few of them. I'll provide links for you, so you can dig in. Also, please note that I'm using Cowork on the $100-a-month Claude Max plan. Cowork will also work on the less expensive $20-a-month Claude Pro plan, but you'll run out of usage allocation faster. 1. Analyze my Home Depot spending The very first thing I tried, back in January when Cowork was just a shiny new beta release, was to have it dig through my Home Depot statements and try to come up with an analysis of what I bought. I fed it a cluster of Home Depot statement PDFs and set it loose. I felt safe-ish doing this because the PDFs didn't show my full account number. Also: Claude Fable 5 is back, but I'm sticking with Opus 4.8 for daily work: 5 reasons why This was only marginally successful, but it wasn't Cowork's fault. The Home Depot statements only included general categories and no full item numbers, so I wasn't easily able to tell how much I spent on tools versus plywood, for example. 2. Organize my PDFs The second thing I did, as documented in the same article I just mentioned, was to have Cowork dig through a copy of my Downloads folder and organize my PDFs. Also: I compared Claude Opus 4.8 with 4.7 in a 10-round honesty test - and a legal prompt broke it This took things a level above Hazel, which is a Mac tool I use to sort files into folders by file type and year. I wanted Cowork to examine the PDFs and divide them into categories. As you can see, it created a solid taxonomy based on the PDFs themselves. It also discovered that many of the PDFs started out with random letters and numbers for names, so it examined them and renamed them to be clearer. I still haven't been brave enough to let this loose in my actual Downloads folder, but I'm reaching the point where it's so good, I just might do that one day. I did like how it categorized everything. 3. Lawyer in a box Back in May, Anthropic introduced its skill set for small businesses. Within that package was an amazingly powerful skill that runs inside Claude Cowork. Once you install it, you trigger it by typing /review-contract. Also: Why I ditched Copilot for Claude in Word, Excel, and PowerPoint - and how you can, too Feed it a contract in Word or PDF form, or as any other doc format Claude reads, and Cowork does a very detailed, very complete, and almost jaw-droppingly excellent analysis of the contract. I ran a few of my old contracts through it, and Cowork found some items I had been worried about and some items that had never surfaced during the original review. My take on this one is that if you ever have a contract, use this. 4. Fixing a stuck server Back in March, I told you about Karakeep, a tool for organizing articles and YouTube videos. It's my locally hosted homelab version of Pocket. For months, it worked just fine. Then it stopped working. The thumbnails that are a signature feature of Karakeep just stopped being generated. The architecture of this is typical homelab complexity. It's a Docker installation, managed by Portainer, running on Linux. Anything could go wrong. Finding it could take days, and my sanity. Also: I tried a Claude Code rival that's local, open source, and completely free - how it went So I set Cowork loose on it. I let it use my browser. I let it connect to Portainer's web interface. I even let it use Webmin to dig around my Linux server. I started the process at 10:38 a.m. Cowork finished at 11:58 a.m. During that time, the AI and I talked back and forth, working out the issue. I can't be sure how long it would have taken me on my own. The fact was, I wasn't on my own. And before noon, a tool I've come to rely on was back up and running. Oh, and the problem was that the Docker user-defined network had no upstream DNS server that it could reliably count on. We fixed that. 5. Aggregating blood pressure readings I've been very intent about optimizing my health for the last few years. In particular, I've been carefully watching what I eat and tracking my blood pressure. A few months ago, I had a doctor's checkup. I wanted to bring him the previous two months of blood pressure history. Also: I let Claude AI control my Mac, and it worked flawlessly - with only two minor issues While I like using my Withings cuff because it records all the readings, exporting those readings is quite difficult. By the time I decided to pull the data, I couldn't find a reliable way to export just the data I needed. My appointment was coming up very quickly. So I just opened the web app, pointed Cowork at it, and told it to dig through it all and build a spreadsheet. It did. I quickly had a chart to present to my doctor. I didn't even have to spend the whole morning cutting and pasting. Good news: My blood pressure is doing great. 6. Sifting through thousands of Gmail messages This is the point at which my trust stretched more than I felt comfortable with, but it was worth it. As I wrote in a recent article, I get a ton of emails from folks who want me to write about them, or who want to talk to me about articles I've written. One week, there was some particularly hot news. I wanted to aggregate some of the comments I had been sent into a sentiment analysis article. Also: I used Claude Code to vibe code a Mac app in 8 hours, but it was more work than magic So, I let Cowork loose in my Gmail. It dug through the more than 7,000 emails I received that week. It found eight that met my criteria. It wasn't just a search on a specific set of strings. Cowork had to read the content of each message for context and decide if it was a fit. I ended up with eight messages I could use in the article. This was a very powerful example of using Cowork as a research assistant. 7. Mitigating a spam attack on my server Of all the Cowork projects I've tried, this is the one that gave me that "we're living in the future" feeling. The quick backstory is that my server was under attack from spammers polluting the site and clogging the server database. My hosting provider reached the point where I was informed that if I didn't stop it, it would shut me down. Also: How to learn Claude Code for free with Anthropic's AI courses - one took me just 20 minutes I had already done some mitigation work with a WordPress plugin that I was working on with OpenAI's Codex agent. Codex is a direct competitor to Claude Code. I use both, but I keep them each on their own projects. The WordPress plugin is a Codex project. But my low-end plan had a fairly limited token allocation for Codex, so I divided up the work. Codex did the coding. Cowork did the analysis and mitigation strategy. This was a brilliant, almost nonstop team effort over a weekend. By Sunday night, the three of us were able to deploy a fix that stopped the spammers in their tracks. We also undid all their damage. Cowork won a spot in my toolbox I am still coming to terms with what Cowork can do, and deciding how much leash I'm willing to give it. Cowork also now has a mobile and web incarnation, which means it can be left to run without sitting on an open laptop. I haven't tried that yet, but I regularly have work that could benefit from some help. I'll probably delegate something to the cloud version of Cowork soon enough. Also: I tested ChatGPT vs. Claude to see which is better - and if it's worth switching Also, note that none of these examples were really coding. Yes, the last example was coding-adjacent, in that it was advising the coding engine in Codex about strategy. But as Anthropic noted in its survey, Cowork seems to be for a lot more than code. I am fairly convinced I'll give this thing more and more to do. So far, the work has involved projects with a definable beginning and end. But with the web version, I may set Cowork to work on a schedule. I might have it do work that might have previously been given to something like OpenClaw, or that I just might have tediously and grudgingly done on my own by hand. Stay tuned. As scary a tool as this thing is, it's proven itself in its first six months to be enormously helpful. I look forward to further helpfulness in future months. That is, of course, if it doesn't murder us all in our beds. Would you use Claude Cowork on your files, your inbox, or your server if it could save you hours of work? Let us know in the comments below. You can follow my day-to-day project updates on social media. Be sure to subscribe to my weekly update newsletter, and follow me on Twitter/X at @DavidGewirtz, on Facebook at Facebook.com/DavidGewirtz, on Instagram at Instagram.com/DavidGewirtz, on Bluesky at @DavidGewirtz.com, and on YouTube at YouTube.com/DavidGewirtzTV.

I tried Claude Cowork on my Gmail inbox after Gemini choked - and it saved me hours of work

Follow ZDNET: Add us as a preferred source on Google. ZDNET's key takeaways * Claude Cowork turned inbox chaos into usable article research. * Gmail search struggled with context and discernment. * Human verification remains essential before using AI-found quotes. How often have you had to sift through a ton of emails to find messages you need? I'm not talking about searching for one or two by a proper name. I'm talking about finding a set of email messages on a specific theme from the thousands in your inbox. I had to do this a few weeks ago. Because of my role as a tech journalist, my inbox overflows with people and companies pitching me on something or other. Because part of my beat is AI, my Gmail gets positively slammed whenever I write about something that a company executive or PR rep wants to be included in. And, you know what? Sometimes I find those pitches to be quite valuable. Also: Anthropic's Claude Cowork heads to the cloud as data shows 90% of sessions aren't for coding Take, for example, the day last month when it became apparent that Anthropic was throttling Fable 5 users down to Opus. A whole bunch of folks, including genuine experts on the topic, sent me emails with their comments on the disclosure. That was a busy week. I received more than 7,000 emails, almost all related to something having to do with AI, Claude, Fable, and the rest. Gmail's Gemini has limits In the past, I would spend hours sifting through my emails to find the information I needed. But this is the age of AI, so I asked Gmail to do it for me. Sometimes, Gmail's Gemini does a pretty good job. Not this time. This time, it failed spectacularly. Here's what I asked it: I think Gmail just couldn't understand the context in each message. It does well when I ask it to find simple messages or topics, or the latest email from a company. But this involved a request for discernment. Gmail wasn't up to the task. But Claude Cowork was. Here's how that works. Letting Claude loose in my email This was the genuinely hard part. Emotionally, that is. If there's anything I consider my confidential crown jewels, it's my email archive. I'm fairly comfortable letting Gemini loose on it. I figure that since it's running on Google infrastructure, Google has probably already hoovered up anything juicy in my Gmail. That ship has sailed. But letting Claude into my email took a leap of faith. I am not good at leaps of faith. So this was hard. But my email has also always been a bit of a bottleneck. If Cowork could save me half a day of tedious mail sifting, it might be worth the trade-off. Also: I let Anthropic's Claude Cowork loose on my files, and it was both brilliant and scary What was it ol' Ben Franklin said? "Those who would give up essential Liberty, to purchase a little temporary Safety, deserve neither Liberty nor Safety." It's a stretch, but letting Claude loose in my email would liberate some of my time, and Anthropic's reputation for security seems to indicate the process would be fairly safe. Hopefully, I'm not giving up either liberty or safety. To make this work, you'll need to connect Claude to Gmail using a connector. Go to Claude settings, then Connectors, and select Gmail. As you can see, I granted Claude permission to read my Gmail, but not write (send) anything. To keep my paranoia properly in check, I disabled the connector once this project was complete. Putting Cowork to work My challenge was twofold: first, make sure that Cowork actually produced the value I needed. Second, don't get so caught up in making Cowork do what I wanted that I spend more time fiddling with it than I would have if I had just gathered what I needed manually. Yes, that's a trap I've fallen into so many times. I'm an engineer. It's a real occupational hazard. I used the same prompt I showed you above. In a little over a minute, Cowork identified 12 PR pitches. Cowork also noticed, "most of these are cybersecurity vendors using the launch as a news hook." Ya think? Also: I had Gemini and Claude write my email replies - but only one sounds like me The good news is that most folks who try to use a news event as a kicking-off point for pitching their own goals generally include some useful perspective relative to the event they're triggering off of. That's what makes them potentially good sources for press mentions. My first exposure to the phenomenon was way back in one of my earliest jobs. My boss, a product manager, was always badgering our PR person to get him "mentions." Basically, he wanted to get his name in the trade press as often as possible. Sure, the company's name and products were important, but his goal was personal promotion. If he could get quoted or mentioned, he'd become better known. This is all part of the weird world of PR. Companies pay PR firms out of their company budgets. But executives in those companies often try to leverage the offerings of their companies and use the associated PR efforts to boost their own careers. And we writers try to leverage the executives and the companies to get more value for our articles. It works because the more perspectives we can present about a news event, the more rounded the coverage becomes. In this case, I specifically asked Cowork to highlight emails that push back against Fable's restrictions as either too restrictive or not restrictive enough. Also: I used Gmail's AI tool to do hours of work for me in 10 minutes - with 3 prompts My next request was for Cowork to sift through those relevant pitches and identify the ones that had enough information to enable me to use them directly in my article. But I also wanted to make sure I had permission to publish, so here was the next prompt. This gave me a set of email messages that might be useful to cull quotes from, and whose message content I knew I had permission to use. Next, I wanted to have Cowork pull good submissions and give me a clear list. But the key was that I didn't want it to modify folks' quotes. Don't worry. "Trust, but verify" is very active here. I built a fairly large prompt in a few stages. First, I set up the loop. It tells Cowork that it should produce a list of entries and do so from the email messages it previously validated. Next is the part that tells Cowork how to format the person's information at the beginning of the entry. I like it this way because I can quickly scan and decide if the person's title or company makes that person more authoritative for the article. Next is the request to extract the meat of the quote. This is the material I considered for working into the article. Notice I specified both word-for-word and no changes. Would it abide by this? Who knows? That last line is key. I wanted the AI to sift through all of my email and cull usable statements, but it's still my responsibility once a quote goes into the article. So if I found a quote in Cowork's list that I wanted to use, I used the link provided to open the email message and personally verify all of the information before I used it. This is critically important. I used Cowork as a research assistant, not as an author. Worth it! I have to say this was a very valuable experience and definitely a tool I will deploy again. Cowork did all the grunt work in a thankless and boring task that normally takes me hours and hours. It sifted through thousands of email messages and presented me with a list of eight validated sources and quotes. If you want to see the article where I used this, here it is: Claude Fable 5 secretly throttled AI researchers, and the internet went wild The time it took for me to do my due diligence with the list of eight sources was just a few minutes. I could devote the bulk of my time to writing, with the bonus of not winding up cranky and stressed because I had just spent the bulk of a day doing glorified clerical work. Also: I tested ChatGPT vs. Claude to see which is better - and if it's worth switching Anthropic calls this "the work around the work," which I think, as a concept, has legs. I will definitely call on Cowork again if I have another administrivia-heavy task I want to offload. Would you let Claude Cowork read your Gmail if it could save you hours of tedious research? Let us know in the comments below. You can follow my day-to-day project updates on social media. Be sure to subscribe to my weekly update newsletter, and follow me on Twitter/X at @DavidGewirtz, on Facebook at Facebook.com/DavidGewirtz, on Instagram at Instagram.com/DavidGewirtz, on Bluesky at @DavidGewirtz.com, and on YouTube at YouTube.com/DavidGewirtzTV.

I was buried in email newsletters. Here's how Claude rescued me

The AI can track unsubscribed emails and mark any re-appearing messages as spam, providing ongoing email management beyond initial cleanup. I'm in the midst of an experiment where I let Claude take charge of my Gmail inbox, allowing it to triage my messages, archive marketing emails, and even draft replies when needed. But while Claude was doing a good job of sifting through my messages, it wasn't cutting down on the sheer volume of email I was receiving. Then my wife had a clever idea. I'd been explaining how I'd set Claude Cowork to label the random newsletters I receive (several of which I don't remember ever having signed up for) as "Archiveable" and then removing the "Inbox" label, allowing me to sort through those messages later. My wife nodded, and then mused that it's a shame Claude couldn't just unsubscribe me from those unwanted emails, too. But wait, I replied: Maybe... it can? The beauty of Claude Cowork, the desktop application feature that lets it perform tasks directly on your system, is that it allows a greater range of creativity than it would have when confined to a chatbox. (Claude Cowork requires a paid Claude subscription; I'm signed up for Claude Pro, which costs $20 a month.) And as we all know, creativity is often required when it comes to unsubscribing from newsletters, especially the more annoying ones. Now, unsubbing from any one email newsletter is no big deal. "Unsubscribe" links are generally easy to find, and Gmail has its own built-in unsubscribe tool. But if you have dozens of newsletters to deal with, getting help from Claude or another AI agent starts sounding a lot more appealing. Going into my Cowork "Morning Gmail triage" project (I'd previously enabled the Gmail connector for Claude), I asked: Looking at the emails in my "archivable" label, which ones are newsletters, and which ones could I safely unsubscribe from? Give me a list but don't actually do it yet. Claude thought for a few minutes (I used the middle-of-the-road Claude Sonnet model for this task, while Claude Opus wrote my initial Gmail triage automation), and then dutifully came back with two main groups of newsletters: editorial newsletters (like the New York Times Cooking email, which I almost never read - sorry, New York Times!), and brand and retail newsletters (including the ones you get signed up for when you create a user account). It also flagged recurring emails (like one from NYC Schools) that I should probably keep. At this point I could have asked Claude to unsubscribe me from, say, all the marketing newsletters in one shot. Instead, I directed it to ask me about each newsletter, one at a time, in multiple-choice format, along with its own recommendation about whether I should unsubscribe or not. Claude began plowing through the list, asking me about each newsletter. It suggested I keep the Film at Lincoln Center newsletter for New York Film Festival updates, for example (I did sign up for that one), but advised nixing Rotten Tomatoes ("low-value listicle content, nothing you can't get by checking the site"). When I was done, Claude got ready for the bulk unsubscribe task, which it would perform using the Claude extension for Chrome. Even better, it wrote itself a JavaScript function that allowed it to quickly find and click the "Unsubscribe" button in Gmail, all without thrashing around in a browser window (and wasting AI tokens in the process). Roughly five minutes later, Claude was done -- 21 unwanted newsletters, 21 unsubscribes. It even suggested a follow-up routine where it would keep track of the newsletters I'd unsubscribed from and mark them as spam if they ever reappeared. Pretty nice, and a great example of AI taking a tiresome chore off my plate.