WEBVTT 00:00:00.000 --> 00:00:02.240 You open your laptop and you type a question. 00:00:02.339 --> 00:00:04.820 You get a quick answer back. Then you just close 00:00:04.820 --> 00:00:07.099 the tab. Right. It's the standard workflow. Most 00:00:07.099 --> 00:00:10.839 of us use AI exactly like this. But that basic 00:00:10.839 --> 00:00:13.679 habit leaves massive value on the table. It really 00:00:13.679 --> 00:00:16.399 does. Welcome to the deep dive. We are doing 00:00:16.399 --> 00:00:18.539 something a bit different today. We're exploring 00:00:18.539 --> 00:00:21.059 a hidden layer beneath the surface. Yeah, we're 00:00:21.059 --> 00:00:24.899 looking at seven specific slash commands. And 00:00:24.899 --> 00:00:27.289 these aren't just minor software shortcuts. They 00:00:27.289 --> 00:00:30.269 transform Claude completely. Exactly. It goes 00:00:30.269 --> 00:00:32.750 from a simple chatbot to a high -performance 00:00:32.750 --> 00:00:35.009 operating system. That transformation is really 00:00:35.009 --> 00:00:37.969 the core idea here. True productivity isn't about 00:00:37.969 --> 00:00:41.289 writing longer prompts anymore. That era is essentially 00:00:41.289 --> 00:00:43.630 over. Right. I mean, it's about choosing the 00:00:43.630 --> 00:00:45.750 precise command for the task. We're going on 00:00:45.750 --> 00:00:48.170 the specific journey today. We're moving from 00:00:48.170 --> 00:00:50.770 heavy reasoning into token efficiency. Yeah. 00:00:50.990 --> 00:00:53.289 Then we'll explore focus management. Finally, 00:00:53.329 --> 00:00:55.350 we'll step into totally hands -free automation. 00:00:55.549 --> 00:00:58.869 I love that progression because typing out endless 00:00:58.869 --> 00:01:01.869 context is just exhausting. Let's start with 00:01:01.869 --> 00:01:04.269 a heavy reasoning aspect. Sometimes you just 00:01:04.269 --> 00:01:06.349 need the model to think harder. That is where 00:01:06.349 --> 00:01:09.170 our first tool comes in. It's the slash ultra 00:01:09.170 --> 00:01:12.430 think command slash ultra think. Right. Typing 00:01:12.430 --> 00:01:15.329 this triggers a massive maximum reasoning mode. 00:01:15.849 --> 00:01:18.609 It gives Claude a thirty one thousand nine hundred 00:01:18.609 --> 00:01:22.379 ninety nine token budget, which Just for context, 00:01:22.579 --> 00:01:25.019 a token budget is the mental scratchpad an AI 00:01:25.019 --> 00:01:28.420 uses to process text. Exactly right. And this 00:01:28.420 --> 00:01:31.379 specific budget is huge. It's about three times 00:01:31.379 --> 00:01:34.019 the standard processing limit. Wow. You actually 00:01:34.019 --> 00:01:36.299 see a rainbow terminal font when it activates. 00:01:36.500 --> 00:01:39.040 Wait, a rainbow font? Yeah, it physically looks 00:01:39.040 --> 00:01:40.920 different on your screen. That sounds kind of 00:01:40.920 --> 00:01:44.340 playful. But the underlying utility is deeply 00:01:44.340 --> 00:01:47.379 serious. It is. You're forcing the AI to catch 00:01:47.379 --> 00:01:51.060 complex edge cases. You're demanding it way difficult 00:01:51.060 --> 00:01:53.500 multi -variable trade -offs. Right, because it's 00:01:53.500 --> 00:01:55.299 not just generating text anymore, it's using 00:01:55.299 --> 00:01:58.099 those tokens to debate itself internally. It 00:01:58.099 --> 00:02:00.640 runs dozens of hypothetical scenarios in the 00:02:00.640 --> 00:02:04.159 background. It actively moves past the easy surface 00:02:04.159 --> 00:02:06.439 -level answers. Let's ground this with a real 00:02:06.439 --> 00:02:10.560 -world example. Imagine you are analyzing a Q3 00:02:10.560 --> 00:02:13.780 newsletter drop. Okay. Your open rate is holding 00:02:13.780 --> 00:02:17.020 steady at 28 percent. That is a very healthy 00:02:17.020 --> 00:02:19.439 open rate. It is. But let's say your click rate 00:02:19.439 --> 00:02:23.259 is dropping. It fell off a cliff from 4 .1 to 00:02:23.259 --> 00:02:26.419 2 .7 percent. Ouch. And this strange drop happened 00:02:26.419 --> 00:02:29.479 over the last eight weeks. Standard Claude usually 00:02:29.479 --> 00:02:31.939 gives very generic advice here. Right. It might 00:02:31.939 --> 00:02:34.710 tell you to write better subject lines, or it 00:02:34.710 --> 00:02:37.349 suggests tweaking the button colors. Because 00:02:37.349 --> 00:02:40.110 standard Claude is optimizing for a fast response. 00:02:40.550 --> 00:02:43.590 Exactly. But slash UltraThink fundamentally acts 00:02:43.590 --> 00:02:46.810 differently. With that massive scratch pad, it 00:02:46.810 --> 00:02:49.229 diagnoses the actual edge cases. It looks at 00:02:49.229 --> 00:02:51.270 your historic sending patterns. Yeah, and it 00:02:51.270 --> 00:02:54.710 ranks the five most likely causes by actual probability. 00:02:55.030 --> 00:02:57.389 And it doesn't just stop at a list. It provides 00:02:57.389 --> 00:03:00.550 one specific A -B test for each cause. You can 00:03:00.550 --> 00:03:02.569 literally run these tests starting in week one. 00:03:02.669 --> 00:03:05.069 It functions as a real diagnostic tool. It's 00:03:05.069 --> 00:03:07.289 like bringing in a specialist consultant instead 00:03:07.289 --> 00:03:08.969 of just asking an intern. That's a great way 00:03:08.969 --> 00:03:11.759 to frame it. It's beat. But this brings up an 00:03:11.759 --> 00:03:14.979 important boundary question. Is this massive 00:03:14.979 --> 00:03:17.840 reasoning budget overkill for my daily email 00:03:17.840 --> 00:03:21.139 drag? Oh, yes. It is absolute overkill. It will 00:03:21.139 --> 00:03:24.620 waste your usage limits instantly. So we shouldn't 00:03:24.620 --> 00:03:26.979 use it for everything? No. We follow the two 00:03:26.979 --> 00:03:29.659 -hour rule for this specific command. If you've 00:03:29.659 --> 00:03:32.120 been stuck on a coding bug for two hours, use 00:03:32.120 --> 00:03:34.520 it. Okay. If you're doing a high stakes legal 00:03:34.520 --> 00:03:37.360 document review, use it. Do not use it for daily 00:03:37.360 --> 00:03:39.780 maintenance or basic formatting. Exactly. Save 00:03:39.780 --> 00:03:41.780 the heavy artillery for high stakes planning, 00:03:42.219 --> 00:03:45.219 not daily typos. Precisely. You want to match 00:03:45.219 --> 00:03:47.879 the compute power to the problem's gravity. Which 00:03:47.879 --> 00:03:50.319 leads to a very natural friction point. Since 00:03:50.319 --> 00:03:52.900 slash UltraThink burns massive tokens, we're 00:03:52.900 --> 00:03:55.039 going to hit our limits fast. Very fast. How 00:03:55.039 --> 00:03:57.039 do we actually balance the budget across a busy 00:03:57.039 --> 00:03:59.789 day? Well, we have to actively trim the fat. 00:04:00.270 --> 00:04:02.469 We need tools that pull in the opposite direction. 00:04:02.569 --> 00:04:04.949 Right. There's an open source skill called slash 00:04:04.949 --> 00:04:07.409 caveman. You just install it once from the community 00:04:07.409 --> 00:04:11.030 skill registry. Slash caveman. Yeah. Its entire 00:04:11.030 --> 00:04:13.110 job is aggressively cutting your output tokens. 00:04:13.449 --> 00:04:17.689 It does this by delivering raw ultra dense answers. 00:04:18.310 --> 00:04:21.560 It strips out absolutely. all the polite fluff. 00:04:21.740 --> 00:04:24.240 Right. It brutally removes those three -line 00:04:24.240 --> 00:04:26.579 friendly introductions. It completely skips the 00:04:26.579 --> 00:04:29.319 repetitive closing summaries. The name is obviously 00:04:29.319 --> 00:04:31.439 a joke about talking like a caveman. Just grunt 00:04:31.439 --> 00:04:33.819 the answer. Yeah. But the useful compression 00:04:33.819 --> 00:04:37.680 is very real and highly efficient. It is. I find 00:04:37.680 --> 00:04:40.199 this great for the messy middle of brainstorming. 00:04:40.579 --> 00:04:43.480 Let's say you need 10 one -sentence LinkedIn 00:04:43.480 --> 00:04:47.360 hooks. OK. The topic is why AI agents fail in 00:04:47.360 --> 00:04:49.810 production. You don't need a greeting. You just 00:04:49.810 --> 00:04:52.129 want the meat of the answer. Exactly. You just 00:04:52.129 --> 00:04:54.589 get a tight, bulleted list. Now you skip it when 00:04:54.589 --> 00:04:56.550 you need polished writing for a client. Sure. 00:04:56.629 --> 00:04:59.250 But for your own daily brainstorming, it is an 00:04:59.250 --> 00:05:01.769 absolute lifesaver. This is especially true for 00:05:01.769 --> 00:05:05.189 our pro plan users. Because that $20 tier hits 00:05:05.189 --> 00:05:08.350 usage limits surprisingly fast. Exactly. So slash 00:05:08.350 --> 00:05:10.529 caveman trims the output. But what about the 00:05:10.529 --> 00:05:13.569 input? Even if the AI's answers are short. My 00:05:13.569 --> 00:05:15.730 sessions still get incredibly bloated. Yeah, 00:05:15.750 --> 00:05:18.029 that happens. I still wrestle with prompt drift 00:05:18.029 --> 00:05:20.910 myself. Two secs silence, I'll be three hours 00:05:20.910 --> 00:05:23.170 in, and suddenly Claude thinks we're coding in 00:05:23.170 --> 00:05:25.730 Python. When you started out writing ad copy. 00:05:25.810 --> 00:05:28.370 Right. It gets messy. It almost always leads 00:05:28.370 --> 00:05:32.290 to strange AI hallucinations. We all fall into 00:05:32.290 --> 00:05:35.420 that trap eventually. Research shows AI models 00:05:35.420 --> 00:05:39.060 heavily lose accuracy as context fills up. This 00:05:39.060 --> 00:05:42.220 is a classic case of context bloat. Which basically 00:05:42.220 --> 00:05:44.620 when a full memory makes the AI forget things. 00:05:45.339 --> 00:05:48.579 Exactly. Under the hood... Claude is forced to 00:05:48.579 --> 00:05:51.420 reread every previous turn. Oh, wow. Yeah, it 00:05:51.420 --> 00:05:53.720 rereads the whole transcript with every new request 00:05:53.720 --> 00:05:55.879 you make. It's incredibly inefficient. It causes 00:05:55.879 --> 00:05:58.540 a major drop in reasoning quality. It also causes 00:05:58.540 --> 00:06:02.079 huge hidden token cost spikes. Right, which is 00:06:02.079 --> 00:06:04.100 why the slash clear command is so essential. 00:06:04.509 --> 00:06:07.269 Typing slash clear simply wipes the current session 00:06:07.269 --> 00:06:09.750 context. It instantly cures that context bloat. 00:06:09.930 --> 00:06:12.230 Consolidating your tasks and periodically clearing 00:06:12.230 --> 00:06:14.490 the slate helps massively. It actually saves 00:06:14.490 --> 00:06:16.970 about 15 percent on your context loading costs. 00:06:17.269 --> 00:06:20.370 The usage rule feels pretty intuitive here. You 00:06:20.370 --> 00:06:23.949 finish a YouTube script, then draft an Instagram 00:06:23.949 --> 00:06:27.550 caption. You clear it. Yep. You analyze med ads, 00:06:27.709 --> 00:06:30.550 them a plan, a newsletter. You clear it. You're 00:06:30.550 --> 00:06:33.269 isolating the variables. So clearing the slate. 00:06:33.890 --> 00:06:37.850 stops the model from mixing up old, irrelevant 00:06:37.850 --> 00:06:40.889 data. Yes, absolutely. If the new task needs 00:06:40.889 --> 00:06:43.410 no previous context, wipe the memory. Right. 00:06:43.610 --> 00:06:45.709 It keeps the processing performance incredibly 00:06:45.709 --> 00:06:48.870 sharp. It stops hallucinations. And it keeps 00:06:48.870 --> 00:06:51.589 your daily operating costs way down. Right. A 00:06:51.589 --> 00:06:53.829 clean slate guarantees faster, sharper, and much 00:06:53.829 --> 00:06:56.110 cheaper answers. It really is the most valuable 00:06:56.110 --> 00:06:58.449 daily habit you can build. OK, so we have deeply 00:06:58.449 --> 00:07:01.540 clean, efficient sessions now. but human focus 00:07:01.540 --> 00:07:04.120 still drifts constantly. Oh, for sure. What happens 00:07:04.120 --> 00:07:05.939 when you get interrupted mid -task? Anthropic 00:07:05.939 --> 00:07:08.199 actually solved this gracefully back in March 00:07:08.199 --> 00:07:11.240 2026. They released a specific command called 00:07:11.240 --> 00:07:13.920 slash B2W. Like by the way. Literally stands 00:07:13.920 --> 00:07:15.779 for by the way. It handles human interruptions 00:07:15.779 --> 00:07:18.240 beautifully. Because focus drift is a huge operational 00:07:18.240 --> 00:07:21.100 problem. Normally you ask a random question during 00:07:21.100 --> 00:07:24.759 a massive task. And the model switches its focus 00:07:24.759 --> 00:07:27.040 entirely. It totally breaks the whole workflow. 00:07:27.240 --> 00:07:31.740 Right. but slash be to do allows safe side inquiries. 00:07:32.019 --> 00:07:34.079 It fundamentally does not break the main tasks 00:07:34.079 --> 00:07:37.879 flow. Let's say Claude is writing a 5 ,000 word 00:07:37.879 --> 00:07:41.339 corporate strategy document. You're pacing around 00:07:41.339 --> 00:07:43.319 and you suddenly remember something important. 00:07:43.579 --> 00:07:46.720 You can just type slash BTW. What is the standard 00:07:46.720 --> 00:07:50.420 image size for a LinkedIn carousel in 2026? Exactly. 00:07:50.959 --> 00:07:54.220 The AI opens a tiny side window execution thread. 00:07:54.509 --> 00:07:57.050 Claude gives you a one -line answer instantly, 00:07:57.490 --> 00:07:59.490 but the main engine continues running safely 00:07:59.490 --> 00:08:01.649 in the background. So it keeps holding the context 00:08:01.649 --> 00:08:03.870 for your strategy document. You maintain your 00:08:03.870 --> 00:08:06.569 complete context integrity. That is legitimately 00:08:06.569 --> 00:08:08.949 amazing. Or if Claude is deeply researching three 00:08:08.949 --> 00:08:11.970 different competitors, you just type slash BTW, 00:08:12.230 --> 00:08:15.209 add the founding year to the final table. Claude 00:08:15.209 --> 00:08:17.189 takes the quick note and keeps researching. No 00:08:17.189 --> 00:08:19.889 hard restart is ever needed. There is zero lost 00:08:19.889 --> 00:08:22.870 context, but there is a very critical constraint 00:08:22.870 --> 00:08:25.829 here. You must never use this for major project 00:08:25.829 --> 00:08:29.209 pivots. Wait, so I couldn't use slash BTW to 00:08:29.209 --> 00:08:31.410 spin up a huge secondary research project? No, 00:08:31.569 --> 00:08:34.389 absolutely not. If your side question is a complex 00:08:34.389 --> 00:08:38.169 10 -minute task... Open a new session. OK. Combining 00:08:38.169 --> 00:08:41.129 two heavy reasoning tasks under one hood dilutes 00:08:41.129 --> 00:08:44.309 the quality. It degrades the accuracy of both 00:08:44.309 --> 00:08:46.870 outputs. Now, keep it to quick questions. Heavy 00:08:46.870 --> 00:08:49.870 tasks need a new session. Precisely. It is a 00:08:49.870 --> 00:08:52.169 quick side window, not a new engine. Mid -roll 00:08:52.169 --> 00:08:54.929 sponsor or read placeholder. So we are managing 00:08:54.929 --> 00:08:57.669 our active open sessions really efficiently now. 00:08:57.710 --> 00:08:59.809 We really are. We're thinking deep. We're cutting 00:08:59.809 --> 00:09:02.649 fluff. We're handling interruptions. Now, how 00:09:02.649 --> 00:09:05.990 do we get the AI to work? when we step away entirely. 00:09:06.129 --> 00:09:08.590 We have to move into the automators. We are finally 00:09:08.590 --> 00:09:10.710 breaking free from the keyboard entirely. Let's 00:09:10.710 --> 00:09:12.909 explore that. The command for this is slash loop. 00:09:13.289 --> 00:09:15.730 This specifically handles local system automation. 00:09:16.330 --> 00:09:19.269 It executes tasks repeatedly at very specific 00:09:19.269 --> 00:09:21.509 intervals. It functions sort of like a highly 00:09:21.509 --> 00:09:23.950 intelligent smart timer. Right. Claude converts 00:09:23.950 --> 00:09:27.090 your plain text timing into a P06 cron expression. 00:09:27.269 --> 00:09:29.570 Which is a standard system for scheduling recurring 00:09:29.570 --> 00:09:33.029 computer tasks. Exactly. It assigns your loop 00:09:33.029 --> 00:09:36.210 a unique 8 -character tracking ID. It's remarkably 00:09:36.210 --> 00:09:38.610 flexible. It supports scheduling by seconds, 00:09:39.070 --> 00:09:41.409 minutes, hours, and days. You can actually run 00:09:41.409 --> 00:09:44.309 up to 50 scheduled tasks per active session, 00:09:44.690 --> 00:09:47.230 though there is a built -in fairness offset. 00:09:47.580 --> 00:09:49.720 What does that fairness offset mean exactly? 00:09:49.840 --> 00:09:52.799 It adds a slight deterministic delay to the trigger 00:09:52.799 --> 00:09:55.779 time. It's usually up to 10 % of the scheduled 00:09:55.779 --> 00:09:58.860 interval. OK. So an hourly task might fire a 00:09:58.860 --> 00:10:01.059 few minutes past the exact hour. Let's see. It's 00:10:01.059 --> 00:10:03.620 like a staggered exit at a stadium. Yeah. If 00:10:03.620 --> 00:10:05.600 everyone leaves at once, the exit's completely 00:10:05.600 --> 00:10:08.440 jammed. This prevents global server spikes on 00:10:08.440 --> 00:10:10.809 the hour mark. That's a perfect analogy. Let's 00:10:10.809 --> 00:10:12.889 look at a highly practical example for this. 00:10:12.929 --> 00:10:15.509 Let's do it. Automating the dreaded blank page 00:10:15.509 --> 00:10:18.230 problem for writers. You simply tell Claude to 00:10:18.230 --> 00:10:20.570 check your local drafts folder every four hours. 00:10:20.950 --> 00:10:24.309 Exactly. It quietly scans your files and picks 00:10:24.309 --> 00:10:27.350 the oldest unfinished piece. Let's say a rough 00:10:27.350 --> 00:10:30.529 draft has been untouched for 11 days. It automatically 00:10:30.529 --> 00:10:33.169 suggests three fresh headline rewrites. And it 00:10:33.169 --> 00:10:36.000 writes one noticeably cleaner opening line. It 00:10:36.000 --> 00:10:38.360 uses psychologically driven headlines instead 00:10:38.360 --> 00:10:40.759 of repetitive social media hooks. It saves the 00:10:40.759 --> 00:10:43.860 entire output as a neat comment block. It even 00:10:43.860 --> 00:10:48.019 suggests a specific 400 word cut. Wow. It diagnoses 00:10:48.019 --> 00:10:51.440 your structural flaws perfectly. But there are 00:10:51.440 --> 00:10:53.179 some strict hardware limitations here, right? 00:10:53.360 --> 00:10:55.659 This relies entirely on your local machine running. 00:10:55.899 --> 00:10:58.899 Yes. Shutting your laptop completely kills the 00:10:58.899 --> 00:11:01.700 active loop. Unfortunately, yes. Closing the 00:11:01.700 --> 00:11:05.159 lid pauses the cron job entirely. It is also 00:11:05.159 --> 00:11:07.500 permanently bound to your current active session. 00:11:07.960 --> 00:11:11.000 Finally, it has a stripped seven day expiration 00:11:11.000 --> 00:11:13.980 date built in. So it's meant for short term sprints. 00:11:14.559 --> 00:11:16.779 It is definitely not permanent. So what is the 00:11:16.779 --> 00:11:20.250 alternative for longer? background tasks. We 00:11:20.250 --> 00:11:22.490 use the slash schedule command. This handles 00:11:22.490 --> 00:11:25.649 pure cloud -based automation. It essentially 00:11:25.649 --> 00:11:28.610 clones your current project state to anthropic 00:11:28.610 --> 00:11:30.990 servers. It runs purely on their remote infrastructure. 00:11:31.250 --> 00:11:33.909 So it fires accurately even if my machine is 00:11:33.909 --> 00:11:36.129 completely turned off. Exactly. It elegantly 00:11:36.129 --> 00:11:38.570 overcomes the main limitation of the local loop 00:11:38.570 --> 00:11:41.279 command. It does. But you have to remember, it 00:11:41.279 --> 00:11:43.940 removes access to your local file system. It 00:11:43.940 --> 00:11:46.740 can't read your hard drive. However, integrations 00:11:46.740 --> 00:11:49.980 with Slack, Notion, and Gmail remain fully functional. 00:11:50.440 --> 00:11:53.419 The daily run limits totally depend on your specific 00:11:53.419 --> 00:11:56.779 subscription plan. Pro users get about five runs 00:11:56.779 --> 00:12:01.000 per day. Max users get around 15 runs. Team tiers 00:12:01.000 --> 00:12:03.840 get roughly 25. Let's discuss setting up a daily 00:12:03.840 --> 00:12:07.039 morning brief. This is arguably the perfect use 00:12:07.039 --> 00:12:09.700 case for slash schedule. You schedule it for 00:12:09.700 --> 00:12:12.919 exactly 7 a .m. daily. It pulls yesterday's top 00:12:12.919 --> 00:12:15.279 news from five different sources. It checks the 00:12:15.279 --> 00:12:18.620 Anthropic Developer Blog and Hacker News. It 00:12:18.620 --> 00:12:21.240 checks the top posts on the AI subreddit and 00:12:21.240 --> 00:12:23.470 TechCrunch. Right. It intelligently picks the 00:12:23.470 --> 00:12:26.269 three most important, relevant stories. It writes 00:12:26.269 --> 00:12:29.309 two concise sentences analyzing each one. Then 00:12:29.309 --> 00:12:31.909 it sends the finalized result to a private Slack 00:12:31.909 --> 00:12:34.269 channel. Whoa, imagine waking up. Your laptop 00:12:34.269 --> 00:12:36.490 was asleep all night, but the cloud just curated 00:12:36.490 --> 00:12:38.789 your entire industry while you slept. Two -sec 00:12:38.789 --> 00:12:42.720 silence. That is a wild feeling. It is a profound 00:12:42.720 --> 00:12:46.080 structural shift in how we work. You read a totally 00:12:46.080 --> 00:12:48.559 clean, personalized brief over your morning coffee. 00:12:49.039 --> 00:12:51.419 The cloud simply does the heavy lifting for you. 00:12:51.620 --> 00:12:54.100 Okay, so if I'm building a system right now, 00:12:54.899 --> 00:12:57.500 how do I instantly decide whether a task needs 00:12:57.500 --> 00:13:01.639 a loop or a schedule? You use a very simple decision 00:13:01.639 --> 00:13:05.159 matrix. Does the task need local files or very 00:13:05.159 --> 00:13:07.039 short intervals? You choose loop. Right. Does 00:13:07.039 --> 00:13:09.879 it run offline or need to run for months? You 00:13:09.879 --> 00:13:12.919 choose schedule. loop for local, short intervals, 00:13:13.299 --> 00:13:15.879 schedule for offline, long -term cloud runs. 00:13:16.039 --> 00:13:18.279 It really is that straightforward. You can even 00:13:18.279 --> 00:13:20.980 build advanced cloud routines later. You can 00:13:20.980 --> 00:13:22.799 actually trigger them dynamically with GitHub 00:13:22.799 --> 00:13:25.059 push events. With all these complex commands 00:13:25.059 --> 00:13:27.679 firing off constantly, how do we track our health? 00:13:27.740 --> 00:13:29.440 That's a great point. We might just be creating 00:13:29.440 --> 00:13:31.779 massive chaos instead of actually saving time. 00:13:31.960 --> 00:13:34.419 That brings us to our final and arguably most 00:13:34.419 --> 00:13:37.179 important command. It's called slash insights. 00:13:38.189 --> 00:13:41.110 Yeah, it functions as your personal AI auditor. 00:13:41.669 --> 00:13:43.909 It securely analyzes your actual session history 00:13:43.909 --> 00:13:47.460 from the last 30 days. It delivers a really comprehensive 00:13:47.460 --> 00:13:50.500 HTML report right in your browser. It visually 00:13:50.500 --> 00:13:53.500 acts as a digital personal coach. Exactly. It 00:13:53.500 --> 00:13:56.360 shows a visual heat map revealing the tasks you 00:13:56.360 --> 00:13:58.799 perform most frequently. Very cool. It flags 00:13:58.799 --> 00:14:01.220 the exact conversational patterns where you are 00:14:01.220 --> 00:14:05.240 wasting tokens unnecessarily. It identifies completely 00:14:05.240 --> 00:14:07.539 unused integrations matching your work style. 00:14:07.980 --> 00:14:10.919 And it doesn't just complain. It suggests highly 00:14:10.919 --> 00:14:14.110 actionable fixes. It might flag that opening 00:14:14.110 --> 00:14:16.269 too many short sessions is getting expensive. 00:14:16.909 --> 00:14:19.830 It will show you how consolidating related tasks 00:14:19.830 --> 00:14:22.090 actually saves you tokens. You really should 00:14:22.090 --> 00:14:23.950 run this on the first Monday of every single 00:14:23.950 --> 00:14:26.350 month. Spend just 10 minutes reviewing the HTML 00:14:26.350 --> 00:14:29.549 report. Identify one small adjustment for your 00:14:29.549 --> 00:14:31.470 daily workflow. But reading a list of problems 00:14:31.470 --> 00:14:34.039 is easy. Changing how you actually work day to 00:14:34.039 --> 00:14:36.600 day is incredibly hard. That is exactly why we 00:14:36.600 --> 00:14:39.080 recommend the seven day mastery plan. Okay. You 00:14:39.080 --> 00:14:41.320 intentionally adopt only one command per day. 00:14:41.740 --> 00:14:44.200 Day one is just getting used to slash clear. 00:14:44.320 --> 00:14:46.840 Use it faithfully between unrelated tasks. Right. 00:14:47.039 --> 00:14:49.840 Day two is slash ultra think. Try applying it 00:14:49.840 --> 00:14:53.029 on one genuinely hard task. Day three is slash 00:14:53.029 --> 00:14:55.389 insights. Read your personal performance report. 00:14:55.529 --> 00:15:00.230 Day four is slash BTW. Use it seamlessly during 00:15:00.230 --> 00:15:02.950 a long writing task. Right. Day five is slash 00:15:02.950 --> 00:15:05.750 caveman. Try it on a messy brainstorming session. 00:15:05.950 --> 00:15:09.190 Day six is slash loop. Automate one simple local 00:15:09.190 --> 00:15:11.769 system check. And day seven is slash schedule. 00:15:12.029 --> 00:15:14.460 Set up your personalized morning brief. I know 00:15:14.460 --> 00:15:15.779 some people are listening right now thinking, 00:15:16.320 --> 00:15:18.340 why not just install and launch all seven of 00:15:18.340 --> 00:15:20.559 these commands this afternoon? Because changing 00:15:20.559 --> 00:15:23.159 your fundamental workflow is psychologically 00:15:23.159 --> 00:15:26.919 very hard. Deploying one new command per day 00:15:26.919 --> 00:15:29.240 prevents it from feeling like massive homework. 00:15:29.340 --> 00:15:32.740 Yeah. It actively prevents complete system overwhelm. 00:15:32.980 --> 00:15:35.899 Avoid overwhelm. Slowly stacking these commands 00:15:35.899 --> 00:15:38.759 one by one builds lasting habits. Let's briefly 00:15:38.759 --> 00:15:41.019 summarize the core philosophy we really uncover 00:15:41.019 --> 00:15:43.580 today. Please do. True productivity in the modern 00:15:43.580 --> 00:15:46.120 AI era is evolving fast. It is no longer about 00:15:46.120 --> 00:15:48.740 writing massive paragraphs along prompts. No. 00:15:48.879 --> 00:15:50.679 It's about deploying the exact right command. 00:15:50.759 --> 00:15:52.879 You match the specific command to the specific 00:15:52.879 --> 00:15:55.980 friction in your workflow. You optimize absolutely 00:15:55.980 --> 00:16:00.340 every second and every token spent. Beat. Thank 00:16:00.340 --> 00:16:02.120 you for joining us on this deep dive. Thanks 00:16:02.120 --> 00:16:04.029 for having me. I want to leave you with a final 00:16:04.029 --> 00:16:07.350 thought to mull over. If Slash Insights can already 00:16:07.350 --> 00:16:10.190 accurately audit our 30 -day habits, how long 00:16:10.190 --> 00:16:12.429 until Claude simply auto -triggers these commands 00:16:12.429 --> 00:16:14.830 for us? That's the real frontier. Imagine it 00:16:14.830 --> 00:16:18.149 realizing we need Slash Caveman or Slash UltraThink 00:16:18.149 --> 00:16:20.850 before we even finish typing the question. AT 00:16:20.850 --> 00:16:21.409 Row music.