Ep 806: Desktop Agent Lingo Simplified: Goals, Loops, Plans, Subagents and how it works in Codex and Claude Code

Talking about prompts and chatbots won't help you talk about AI strategy in 2026. 

You've gotta know the ins and outs of loops, plans, goals, subagents and more. 

In this episode of Everyday AI, we're breaking down the agent lingo and how the key terms play out in systems like Codex and Claude Desktop. 

Desktop Agent Lingo Simplified: Goals, Loops, Plans, Subagents and how it works in Codex and Claude Code -- An Everyday AI Chat with Jordan Wilson

Newsletter: Sign up for our free daily newsletter
More on this Episode: Episode Page
Today's Episode on LinkedIn: Thoughts on this? Join the convo on LinkedIn and connect with other AI leaders.

Upcoming Episodes: Check out the upcoming Everyday AI Livestream lineup
Website: YourEverydayAI.com
Email The Show: info@youreverydayai.com
Connect with Jordan on LinkedIn

Topics Covered in This Episode:

1. Desktop Agent Vocabulary Primer
2. Agent Harnesses: Codex vs. Claude Code
3. Desktop Agent Plans: Features and Workflow
4. Goal Setting in Codex and Claude Desktop
5. Plan vs. Goal: Key Differences
6. Agent Loops: Automation and Verification
7. Sub Agents: Parallel Task Management
8. Context Windows and Task Delegation
9. Guardrails, Verification, and Cost Control
10. Transition from Chatbots to Autonomous Agents

Timestamps:

00:00 Shifting focus to AI agents

03:28 Accessing the Start Here series

09:31 Using plan mode in clawed desktop

12:04 Understanding plan vs. goal mode

14:25 Setting project goals and planning

19:33 Accessing Start Here series

22:03 Building effective training loops

26:48 Managing sub agents effectively

27:30 Setting up sub-agent system

30:47 Closing and subscription reminder

Keywords: 

desktop agent, desktop AI agent, agent lingo, agent vocabulary, long running agent, autonomous agent, codex, Claude Code, Claude desktop, AI harness, agentic harness, agentic tools, super app, Microsoft super app, OpenAI codex, long running desktop agents, plan mode, planning phase, agent plan, goal setting, AI goal, agent goals, loop mode, agent loops, scheduled automations, sub agents, agent subagents, context windows, parallel work, context hygiene, verification steps, approval points, skills, automations, API token usage, project threads, co work tab, code tab, work trees, checkpoints, file access, browser automation, human in the loop, token efficiency, agent delegation, AI supervision, knowledge work automation, AI subagent management, desktop agent mental model, computer control, AI project management, AI workload delegation, remote steering, front end chatbot, proactive AI, AI context sharing.

Start Here ▶️

Not sure where to start when it comes to AI? Start with our Start Here Series. You can listen to the first drop -- Episode 691 -- or get free access to our Inner Cricle community and all episodes: StartHereSeries.com 

Also, here's a link to the entire series on a Spotify playlist. 

Transcript

SPEAKER_00 0:02

Remember back in 2024 when knowing how to use a custom GPT was a differentiator? Or in 2025, when knowing the difference between a skill and a project and Claude could be a competitive advantage for your company's AI efforts. Well, those days are gone, and so is most of the useful lingo because controlling front-end chatbots and models and modes are now table stakes. The differentiator now is controlling long-running desktop agents. And it sounds easy in theory until you realize that the language you've been building up on the front-end AI chatbot era and the language of the long-running autonomous agents are not exactly compatible. So today we're giving you the primer and establishing some baseline vocabulary and concepts you'll need heading into 2027 when long-running AI agents become the new norm. So here is the big picture. Agent vocabulary is now an essential skill set. And yeah, it's changing all the time, which is one of the reasons why we do this thing every day. But agents can obviously read and write files, they can run tools, call their own, create their own apps and plugins, they can fix mistakes and work for hours unattended, which is both a good and a bad thing, depending on uh how active you are in your agents. And every new term now names a problem you only hit once an agent run runs long. Uh right. So so so much of the previous uh terminology that we use with AI chatbots, right? You had an instant feedback loop for yourself, whether it worked or not, and that is kind of gone. You really have to be paying attention. And not knowing the words now, well, means you might set a vague goal, you might put up weak guardrails, and that run that you think might fix a problem as you go take a walk, might not really go anywhere. And learning this new agenc language is becoming as essential as learning how to prompt, as that was helpful. Uh, so on today's show, we're going to learn what loops, goals, plans, and subagents actually mean without the jargon. Uh, we're gonna go over how all of these terms work in Codex and Claw Desktop so you can know how these features work together. Uh, you're gonna learn why fluency in these terms is the skill separating operators from spectators, and you're gonna know the mental model that makes the whole vocabulary click starting now. All right, if you're new here, welcome. This is the Start Here series. The Start Here series is part of everyday AI's ongoing effort to give whether you're a new listener or a seasoned AI expert to give everyone an essential podcast series to both learn the AI basics and double down on your AI knowledge. So if that's what you're trying to do, sweet, me too. Let's do it together. Make sure to go to starthirseries.com. That's going to give you exclusive access to our inner circle community, and you're going to be redirected straight into our Start Here Series space, which has a playlist uh for all of these episodes, all in order. So it's easy to uh go through them all, as well as um you can read about them all on the page and connect with other people who are along the journey with you. So if you missed our last Start Here series, that was volume 29, where we talked about the open source surge and if models like GLM 5.2 uh make open source an enterprise priority. But today we are talking about the desktop agent lingo simplified. So first let's zoom out completely. All right. So if you are not someone that's using codex or clawed desktop or anti-gravity or cursor, uh, some of this might not make a ton of sense. And that's okay. And maybe this show is more for you than anyone else. Uh, because if you are using something like Codex or uh Claw Desktop every single day for hours, this episode will probably be a review at best, but I think still helpful. Uh so if you're like, okay, I don't use these tools, no, you need to listen up uh and you should start using them. Uh, but I want to talk about the shift, obviously, from the uh AI chatbot that's just very reactive versus the proactive autonomous desktop worker. Um, and really what separates them is the harness, right? So uh sometimes when you talk about a model, you talk a lot about the harness or where it lives and how it accesses how it how they access all of these tools, uh right. So let's just use an example codex from OpenAI, uh, right? It's my favorite harness. It's the one I use most, but a lot of people don't even know. You can use other models inside codecs. Codex is the harness, uh, right? You can't do that with Claude Code, you can do that with anti-gravity, codecs and some others, cursor as well. Uh, so the harness is how all of these uh agenc tools come together and they can work over long term. Uh right. When you see all these stories, you're like, oh, this person had an agent running over the weekend or overnight. Well, that's really made possible by the harnesses that these companies create that give the models and the tools essentially a sandbox to do all this. And, you know, it's kind of uh today's episode is really a primer on understanding what happens under this, the hood of the harness, so to speak. So we're not gonna get super technical. This is uh for beginners and maybe intermediate people as well. But that's what we're truly trying to understand. And like I said, even if you're not using these desktop tools now, I think most people will be using them uh come 2027, right? We've heard from Microsoft that they're coming out with their super app, uh, right? We tackled uh super apps uh on the show and on the start here series. Uh, what episode was that? Bringing that up here. That was our episode uh 799 or volume 28 of the Start Here series. So, you know, Microsoft is bringing out a super app. So these super apps uh or these agentic harnesses, right? If you want to get a more technical term, essentially they are a much more powerful version of a web-based chatbot that can run autonomously on your computer. Uh, you can run them in loops, you can run goals. We're gonna go over all these things, you can schedule automations. So the big difference is well, it can act autonomously, it can share the context uh across different chats. Um, and it can read and write uh to your computer. So any file, just like a human really would, it can use your actual computer with computer use, it can use your browsers, right? Uh, has built-in browsers. So uh, you know, definitely go listen to the uh AI super apps episode uh if you want a little bit more primer, but let's get straight into it now. So let's talk about plans. And I think plans are actually one of the more underrated features of these harnesses, specifically in Claude Code and in Codecs, right? Uh every once in a while I'll ask people how they're using uh these different harnesses. And I people don't really talk about or use plans as much as they should. So a plan essentially just reveals the route before an agent asks uh or acts. So a plan shows the intended steps before, you know, the files, the tools, the app changes, all of those things, and planning exposes the assumptions, uh, right. So likely files, approval points, and verification steps. So plan modes matter because desktop agents can change work fast and they can work for a very long time. Uh, right. So it's kind of like uh a blueprint for a building, right? You wouldn't just, if you had unlimited resources and you wanted to build a building, you wouldn't just go to someone and be like, hey, go build a building. Or if you're building a custom house, you wouldn't just be like, hey, I want a house and make it awesome. Uh, right. You'd probably sit down with an architect and you would go over the blueprints, the floor plans, uh, zoning restrictions, requirements, all those things, right? It's might sound tedious, but you're probably gonna get a much better result in the long run if you sit down and have the conversation with said architect or uh general contractor, right? Whatever it is. That's the same thing a plan is. It is literally a plan. Uh so a lot of people will usually just point uh, you know, clawed desktop or codex uh to a folder or share a little bit of context and say, get to work, buddy. Uh, not a good idea. So the plans are essential. So codex, right, uses uh the kind of plan, pair, and execute as collaboration gears. Uh so it reads, it analyzes, and then it proposes things to you, and then it waits for you to approve it before implementation. Uh so similarly, uh, that's how Claude Desktop works. Uh, it actually works out nicely that these work very similarly in Claude Desktop and in Codex. Uh so if you're actually using these and if you want to see like how does this work, where's the button, right? Codex has a plan button. I don't believe Claude Desktop has it, but you can invoke them each the same way, which is just a backslash and then a plan, uh, right. And then they'll kind of walk you through and then you will approve the plan. Uh so in the same way, how I am a very uh adamant telling people like, make sure you read the chain of thought after you've gone through uh, you know, and gotten an output out of a large language model. I am that much um in favor of using plan mode. It is the equivalent on the front end. Uh, right. So if you've never used these harnesses, think of something like Google's uh Google Gemini's deep research actually does a really good job of this, right? Uh before you go out and do a deep research with Google Gemini, it actually gives you a plan to approve so you can see it and if something's wrong, you can modify it. And this is really important because when we talk about the the shift from token maxing to token efficiency, right? Depending on how you set these agents up and what kind of task you're giving them, they might run for a long time. And you know, if you are using these on a company plan, chances are you're paying via the API. Uh so not having a plan in the same way telling uh a builder to build you an amazing house can be a very expensive step to overlook. Plan plan mode, you have to think. It's not slowing you down. A lot of people are just like, I want to build, I want to break stuff, I want to burn tokens. No, it's a guardrail, right? Not a slowdown. So planning separates thinking from doing, uh, but it's not, you know, it's not enough because risky work still needs, right? That read-only access, the work trees, the checkpoints. And then once your route is approved, you know, then the permissions kind of define the real blast radius. Uh, so don't just rush into giving a long-running agent uh keys to the castle or you know, giving it access to just run for hours and burn through your API bill. All right, next, let's talk about goals. All right. And I I kind of put these in an order that I think might make sense for most. Um, I generally will start with a plan. And the way I actually do it is I use a plan and then uh parlay uh the findings of that into a goal. So it's a little confusing, and uh Claude Code and Codex work a little bit differently. Uh uh in for plan mode, it will still work through your plan, but a lot of times it will stop once it's gone through the steps. Uh so the biggest difference between a plan and a goal is on the front end, uh, a plan literally just outlines it and you can see uh codecs or claude code work through each step, and there is a visual indicator, which is great. Goal is a little bit different. Goal is you literally give it an end goal and it will not stop and sometimes loop until it hits that goal, right? So uh a lot of times you have to be careful using goal if you don't go through a plan mode first, or if you don't go through an old school best practice prompt engineering, context engineering, like we used to teach with Prime Prompt Polish, to really share that context and have an understanding with the model of what ultimately the output looks like. So if you just blindly, and a lot of people do this, right? They just blindly throw a bunch of context at Codecs or Claude Desktop and they give it a goal and it will keep going until it hits that goal. And if that goal is maybe unattainable, uh, yeah, that's where you can have it work for uh hours on end, or maybe a day or longer, and all of a sudden, yeah, your uh your your bills through the roof. So uh let's talk a little bit about goals. So it's really defining the finish line before you start any motion. And like I said, I always go through a planning phase first. I usually make the plan kind of big uh to work through it, uh, let you know, Codex or uh Claude go through and do some work. Then I'll see, you know, there's always gonna be shortcomings the first time you give something uh something to someone. So I'll see where it went wrong. I'll go through, read the summarized chain of thought, then I will kind of reprompt it and then make that as a goal. All right. And sometimes I'll literally just copy and paste the original plan, make some modifications, obviously, because some things on the list or on the steps of the goals are gonna get crossed off. Some things aren't. Uh so I usually do a lot of manual nitpicking, uh, transitioning from a uh a plan to the goal. But the goal is the outcome and agent checks across the long run. And strong goals specify the audience, deliverable source material in the done condition, which is why I think it is usually best in general terms for you know, general knowledge work. And again, I'm coming at this from like a general knowledge work. Yes, I do some coding software dev stuff, building myself uh, you know, cool projects and uh uh the uh products, extensions, all those things, but I am talking about this through general knowledge work. So even if you're not, you know, writing software or you know, vibe coding something, let's just say your vibe working, I still think that this holds true. Um so weak goals make plans, loops, and subagents drift quietly, which I think it's important to go through, especially if you're talking about uh a project that you want to do, right? If you're just trying to knock off a uh a task that might normally take you or a single agent on the web, you know, five, 10 minutes to go through. I don't think you necessarily have to go through the plan and the goal in combination. But if you're talking about knocking out an actual project, which is what these systems are capable of doing right now, that might take a human two, three, four, five hours or two, three, four, five days. That's when I think it uh, you know, when you're talking about your ROI and what you have to invest on the front end. This is one of those things where you're gonna have to build a bridge in order to save the four days driving around the mountain. Uh, so goals differ a little bit uh across Codecs and Claude. So Codec's goals are a little bit more persistent and editable and also visual in the composer. There's a nice little toggle uh button in the goal first threads uh in codecs, kind of keep the objective across turns and sessions. Uh Claude code uh via the desktop version, not the uh CLI. It does still support the backslash goal in the same way. Uh, but uh cowork is a little bit differently. Uh it works a little bit different. That's the other thing. People uh, you know, and for me, I personally use uh, you know, codecs, and I do have to tell people this because if not, it would be extremely irresponsible of me. Uh, you know, the big difference between codex and claw desktop is clawed desktop is fragmented. There is a chat, a co-work, and a code tab. And those tabs have no clue what the other is doing. So there's, you know, uh certain things that work really well in cowork goal, uh, such as the visual indicators that don't always work as well in the code version of goal. So, you know, even as we go over individual features, uh, because clawed code by default, uh, or sorry, clawed desktop by default is siloed uh between the chat, the co-work uh and the code, you know, even things like goal work a little bit differently uh in co-work and in clawed code. All right, another thing to keep in mind um a goal is not just a step-by-step prompt, right? So over-specified steps can actually fight the agent's own planning process. So you don't have to know everything going into it, right? You don't have to like be like, oh my gosh, I can't use this plan thing and the goal thing because I don't have experience. I'm not a software engineer. Absolutely not, right? You have to be able to communicate and understand what you want and then work with an agent to you know go through the steps to get there. But under specified outcomes, make the agent invent missing like success criteria. And that's what you absolutely want to avoid. All right. So now let's talk about loops, right? So loops are kind of the uh new-ish trend, uh, right, even though we had the Ralph loops, you know, many, many quarters ago. Uh, now loops are making their uh viral reappearance again. So loops uh essentially can turn plans, approved plans, into agent work. Uh right. So similarly, like a heartbeat if you're using open claw, uh, right. So think of it like this it's it's something that you can schedule to happen over and over and over and over again. So a loop means you you know the agent is going to observe, plan, act, check, adjust, and repeat. It is literally a loop. Uh, so a chatbot would run once, while a long-running agent can loop many times if you give it that instruction. So the loop only becomes useful when the agent verifies each step. Otherwise, you're just burning tokens like you're still trying to climb the uh internal meta um, you know, token burning leaderboard. So uh, you know, loop review looks a little bit different in each product. So codex uh runs every task loop inside a project organized thread. Uh so you know, you can also uh build loops as skills or automations, right? So you can just tell them, hey, here's what I want you to do, right? Every, you know, every hour I want you to go through, you know, triage my email, my calendar, uh, and my drive, uh, you know, respond to any emails that I might want to, update any decks, and I want you to do this every hour. I want you to do this, you know, every day. Uh, right. So that's an example of a very oversimplified example of what could be considered a loop. Right. So Claude Cowork, again, a little bit different. It shows the steps with citations to the files and the messages, which I really like. I like that uh that piece in co-work uh a little bit more than I like it in Claude Code because it has that dedicated side panel on the right. So even if you are running technically a loop, you can see those steps visually get checked off. Uh Claude Code surfaces the plans, the tasks, the sub agents, the diffs in the progress panels. The biggest thing to talk about when uh you know talking about loops is having very clear verification, uh, right? Whether you want to run something uh a first time, right? I would literally, and I've done this before, you can work through, uh just go through a natural language, work with codecs, work with Claude Code, and say, hey, I want to create a loop. I want to ultimately save this as a skill earn automation, uh, or you know, usually both save it as a skill and an automation that you can schedule and run. But say, hey, instead of giving you the full loop, I want to work with you uh through this. Here, you know, here's ultimately what I want to do. I want to, you know, check this website uh, you know, twice a day. Uh, you know, this is my whole company, we live and die by this website, right? It could be industry news, could be uh I don't know, stocks, finances, um, etc. Right. But think if if if you are someone that has to pay very close attention to the market, something in healthcare, I don't know, FDA regulations, and there's all these things pinging all the time. Uh, right. Maybe you want to create a loop that you train or you help build a skill set with codecs or clawed code that goes in there every so often. You know, you see what's new, you run it through your context, your decision-making process, etc. Work through those steps in the loop one by one. So the um, so this harness or the model in the harness can understand what verification looks like, what success looks like at each step of the loop. If not, bad loops, bad, right? Bad loops turn into producing uh, you know, overly polished work that just is maybe wrong. Because when one loop gets overloaded, then sub agents are maybe going to split the work. All right. So loops are great. You can save them as a skill in automation. Think of it similar to like a heartbeat. Uh, if you've used OpenClaw, the great thing, all it takes is natural language, uh, right. I do think in this aspect, codex is a little bit better uh at setting these up. I think that they're uh because not being fragmented, uh, it's a little bit easier to just chat uh with codecs in natural language to set up those loops. All right, and then last but not least, we have sub agents. So all right, sub agents like loops are where things get very interesting and potentially dangerous if you are not uh being hands on. So if you're being uh uh laissez faire. Human in the loop and setting all these things off. I wouldn't recommend going too heavy and the pains on uh loops and subagents. It will get expensive quickly. If you are very hands-on, uh, I think loops and subagents are great. So here's what subagents are. Well, they're just helper agents with focused assignments and separate context windows. So they help with parallel work, not just vague requests to think harder. The real value, though, is context hygiene before the consolidation begins. So subagents control um kind of uh a different aspect of a job. So one easy thing, right? Uh an example of how I use subagents uh in codecs uh and in Claude. So you can just use them um by invoking them. Say use sub agents for this. A lot of people are like, How do you use sub agents? Well, you natural language, use sub agents for this. So, as an example, one thing I like to do is I like to build with codecs. All right. So I build with codecs. A lot of people are like, Oh, Claude is better at front end. Yeah, it's way better at front end, but I don't know. Uh, codecs has this thing called the most powerful AI um image model in the world. So start with the uh AI image gen in codecs, uh, use the front end skill, takes three seconds, and you generally will have a much better front end uh if you are building, you know, a piece of software that's just what I'm using as an example, right? If you use the image gen in the front end design skill, you're gonna have way better than what you would get out of Claude Code if you do that. If not, obviously uh Cloud Code is gonna be better. But one reason or one way I use subagents, uh I I love Claude's ability to uh really investigate and use subagents across multiple things. So I'll sometimes I'll assign, you know, I'll say, hey Claude, I want you to do these sub agents. I want you to look at, you know, one, look at the entire uh repo. Number, you know, two, look at the feature set, three, look at the language, right? So then each subagent uh can really have a more defined and refined um kind of uh task, right? So it's the same thing. If you just walked into a room of 20 employees and they're there to help you, and you say, Hey, 20 employees, uh, go help me with this project. Here's where we're at, have fun, uh, right. They might decide upon themselves, okay, let's split this up. But if you say, hey, designer, go look at the design, hey copywriter, go look at the copy, uh, hey engineer, go look at the, I don't know, the security on the back end, make sure everything's you know tightened up. So all you have to do is well say, go use sub agents. Obviously, these models are smart enough where you can just say that broadly, um, and they will uh, you know, usually uh, you know, assign one sub agent to a specific task. Uh, usually I'll just broadly say go use sub agents and poke holes in ABC. And then I'll see exactly how they did it. Uh Claude and Codecs allow you to click uh onto a sub agent and see exactly what they're working on in real time, which is cool. I love how Codecs names them. Uh the Claude uh names, I forgot how they name them, but um, you know, usually I'll do a quick general subagent run if it is a big project that's super important, right? That I can spend kind of the token budget on. Uh, I'll see what they did, what went wrong, what didn't, uh, you know, how it can be improved. And then I'll probably assign roles the second time I have another group of subagents do it. Sometimes I'll have sub agents work on the front end before the work actually starts to scope it and make sure it makes sense on where they're spending their time. Sometimes I'll have them work on the back end after the work is done, or a combination of the both. And the best thing is well, you can just tell Codex or Claude, like, hey, have sub-agents work on the front end before you start this project to make sure everything is correct. Uh, then go do the work, then have a separate group of sub-agents really tear it apart on the back end, right? That's all you really need to say, and they will do that. It is kind of like this having this parallel work stream that can disagree, duplicate, uh, you know, miss shared constraints. And that's great. And then the main agent can compare findings, resolve conflicts, synthesize. I've even set up a skill that has like a three-tier uh kind of uh agentic management system. So I don't even have to go through and type this, right? So if I know a project is pretty big, or if I'm gonna be working, you know, it's like, hey, I'm setting this off to work overnight, uh, it's it's an ongoing project, and I'm not gonna go through the the planning and goal. I know I have this, you know, subagent uh system where I, you know, say, hey, this one's gonna go in and be a contrary uh or uh contrarian and you know, tear apart all this and second guess every single thing that we do before it goes to production, whatever it may be. Uh so that's just uh a way to go through subagents. All right. So now as we wrap up, the real constraint here is actually your machine. This is the big thing to keep in mind because your machine has to be on. It has to be uh right, you have to have these programs open. So yeah, you can do all these great things like control your computer, right? And and control the browser. And think of that as well as we talk about uh, you know, the desktop lingo is it's literally as if you were talking to a human sitting in front of the computer, but it has to be on, right? A sleeping laptop can break remote steering. Uh, you know, if you run out of power, if your internet goes off, everything, well, from an agentic perspective stops. Where that's a little bit different. The advantage of using on the front end chatbot, uh, those you know, singular tasks, those singular one-off prompts can continue to go on. But the beginner mistake is not managing anything, uh, right? The beginner mistake is being like, wow, these things are super powerful, these agents. I can just go in and you know, drop it some context and say, go do this work and wow, look at it. You're gonna get just slop work almost every time, uh, right. Because also bad handoffs say research this, organize that, make that better, and then you just use the output. Uh right. Good handoffs are well defining the goal, the plan, the permissions, the workflow, the verification, the subagents, the loops, uh, right. That's the differentiator now, not just saying, oh, I use codecs, oh, I use Claude Code Desktop, right? Uh, the handoff turns the agent lingo into repeatable desktop delegation. So, like I said, the new skill is number one, understanding how to talk to your agents, but then using those skills to supervise work, not just prompt, right? The agentic layer is growing thicker and thicker by the day. That doesn't mean the human layer, uh, both uh, you know, I call it the uh, you know, the agentic human sandwich, right? We are the bun. We are providing uh the input and the context on the front end and then ultimately the verification. But that doesn't mean we're hands-off in the agentic layer. We have to constantly be monitoring, uh improving. So we don't just, you know, blindly let these agents loop and uh, you know, blindly assign work to sub agents, uh, right? We can just spend more time uh as the bun and far less time as the meat, to use my old saying there. All right. I hope this one was helpful going over the basics of desktop agent lingo, simplifying goals, loops, plans, uh, and sub agents, and a little bit of how they work in to in codex and in clawed code on the desktop. So if this was helpful, please go to starthirseries.com. If you haven't already, make sure to subscribe to the podcast. I would really appreciate that. Thanks for tuning in. Hope to see you back tomorrow and every day for more everyday AI. Thanks, y'all.