00:00:00The creator of Claude Code previously revealed his entire workflow and we already made a video on that.
00:00:04We took a lot from it and started applying his principles into our own work.
00:00:08But we wanted to go further so we started looking for the specific skills and tools that his team actually uses daily.
00:00:14We went through their posts, the open source repos, the official plugin marketplace and pulled out the skills and commands that they actually use.
00:00:21Some of these you can install right now, others are internal tools that the team never released but we figured out what they do and how you can build your own.
00:00:28Now one skill that they use is the Frontend Designs plugin for Claude Code.
00:00:32It was developed to convert designs using simplistic prompts while enhancing them in terms of UI and UX.
00:00:38It is available open source and the repo contains a skill with specialized instructions for design.
00:00:43This is available upfront in the official Claude Code plugin repo.
00:00:47Basically the reason for this is so that AI can avoid general AI aesthetics, which are the usual designs that models tend to converge toward most of the time.
00:00:55The installation process is simple, you can just add it from the Claude Code plugin's official marketplace.
00:01:00But if it is not installed, you can install it using the plugin's add marketplace command and then install it from there.
00:01:05After a reload, the plugin will be available for use.
00:01:08Now once it's installed, you can invoke the skill using the slash command directly and specify what you want to do with it.
00:01:13Since skills are usually invoked whenever Claude feels the need for their functionality, this skill will also be triggered if Claude determines it is necessary and it can invoke it during execution as well.
00:01:24Now another skill that the team behind Claude Code uses daily is the batch skill.
00:01:28This skill is integrated directly into the product and is designed to automate tasks that can be easily parallelized such as migrating code from one library to another or performing similar operations.
00:01:39You can use it directly inside Claude Code and give it any instruction after the batch command to execute your task.
00:01:45Once you provide Claude Code with that prompt, it first identifies the task and breaks it down into subtasks by entering plan mode.
00:01:52This allows it to explore and plan the implementation thoroughly before executing.
00:01:56Now you might think that Claude's agent system already breaks down tasks into smaller parts and executes them in parallel.
00:02:02So why is this necessary?
00:02:04But the difference is that the batch command uses separate work trees which are isolated copies of your repository letting each agent work in its own environment.
00:02:12Therefore, this prevents agents from affecting each other's work.
00:02:15The plan generated during planning contains the current state of the app,
00:02:18a division of all the work into units, a list of required additions, and instructions for how end-to-end verification should be performed.
00:02:26Once you approve this plan, it will spawn the same number of agents as there are units of work,
00:02:30giving each a dedicated prompt and ensuring that each agent operates in a separate work tree.
00:02:35Each agent reports its progress back to the main agent, and once all of them have completed their tasks, the main agent merges all the isolated work trees into the main branch.
00:02:44It can even manage the pull request if a remote is configured.
00:02:48Now they also use the code simplifier plugin frequently.
00:02:51This plugin is open source and can be downloaded for use in your own project.
00:02:55It's available in the official plugin's repo and can be installed following the same installation steps.
00:03:00Once installed, you'll see it in your list of installed plugins and you can also search for and install it directly from the repo.
00:03:06This plugin refines code for clarity while preserving functionality.
00:03:10To use it, you provide a prompt and it initializes the agent inside the code simplifier.
00:03:15It then spawns a single agent to simplify the entire code base by removing duplicates and unnecessary files, making the code base cleaner and easier to manage.
00:03:23Once it completes its work, it returns a summary of the changes it made.
00:03:27But this is actually similar to the Claude Code Simplify skill which was released as a product within Claude Code.
00:03:32The main difference is that Simplify spawns three agents and rigorously evaluates the code across multiple metrics to make it simpler and more effective.
00:03:40So Simplify is a more thorough version of the Code Simplifier plugin and both tools are heavily used by the team.
00:03:45But before we move forwards, let's have a word by our sponsor.
00:03:48Airtop. If you are spending hours every week manually doing repetitive tasks, there's a much better way to handle that.
00:03:54Airtop lets you build AI agents that run 24/7 in real cloud browsers using plain English.
00:03:59No code, no APIs.
00:04:01I set up their competitor spy template, gave it three competitor URLs and told it to monitor them daily for pricing changes, new features and positioning updates.
00:04:10Every morning, Airtop spins up a cloud browser, navigates each site, handles captures and anti-bot measures, compares everything to yesterday's data and drops a summary straight into my slack.
00:04:20Set it up once, 10 minutes and now I get a competitive intelligence briefing every morning before I even open my laptop.
00:04:26Build reliable automations in minutes and sign up for Airtop using Code AI Labs for a free month of their starter plan.
00:04:32Click the link in the pinned comment and build automation with Airtop.
00:04:36Now another skill that the team uses internally is the Verify skill.
00:04:39It is not mentioned in their tweets but from the leaked source code we can see that it is an internal skill for Anthropic team members, hidden behind a CLI flag and not available to others.
00:04:49This skill verifies code changes by running the app, testing the changes from different angles and automatically fixing failures.
00:04:55Since this is used internally, the system prompt is not inside the product's code.
00:04:59It is injected directly into the environment when any one of the team is actually using it.
00:05:04This may be why it hasn't been released publicly as it is highly project specific.
00:05:08But if you are building your own project, you can create your own Verify skill using the Skill Creator or by using the repo as a template.
00:05:15It will scan your code base and generate a Verify skill tailored to your project.
00:05:19Verification can look different depending on the project.
00:05:22Some might use Playwright, MCP verification, tests or other methods.
00:05:26But the core idea is that the agent need to have a way to confirm its changes are working as intended and this skill handles that heavy lifting.
00:05:33The skill's instructions include step-by-step guidance on how to handle code changes and report back using the available tools.
00:05:39It also contains examples in CLI.md listing all CLI tools the skill can access, including linters, exit codes, NPM test files and more.
00:05:49For optimal use, it should be configured with test cases and visual verification using the Claude Chrome extension.
00:05:55Now another internal skill hidden inside Claude code and available only to Anthropic team members is Skillify.
00:06:01This skill captures an entire session and turns it into a reusable skill.
00:06:05Like the Verify skill, it is protected by an environment flag.
00:06:08But unlike Verify, the system prompt for Skillify is included in the source code.
00:06:12The purpose of Skillify is to record a session where you brainstorm a workflow with Claude, refine the process through multiple conversations and then save it as a reusable skill.
00:06:21It identifies repeatable processes discussed, all tools and permissions mentioned, the agents involved and everything needed to create a skill.
00:06:28It then interacts with the user to clarify intentions, breaks down the steps, finalizes the skill.md file and generates a detailed guide containing all necessary instructions.
00:06:38But you don't need access to the internal version. You can create your own Skillify skill using the source code as a guide.
00:06:44Once invoked, the Skillify command thoroughly analyzes the session and confirms with you whether its deductions are correct.
00:06:50It asks follow-up questions to clarify the workflow, then generates a skill.md file and provides it for your review.
00:06:56After confirming that it matches your needs, you can save it to your project and use it as a reusable skill.
00:07:02Now another tool they use is the Claude Code Security Scan command which scans the codebase for vulnerabilities and suggests patches based on various factors.
00:07:10It has been used internally by the team to identify multiple security issues and was later released as a security scan feature inside Claude Code.
00:07:17The system prompt can be accessed from the leaked source, and the tool is configured to perform security scans from multiple angles, including input validation, authentication issues, secret management, injection and code execution, endpoint exposure and more.
00:07:31It also includes an in-depth analysis methodology and incorporates multiple standards directly into the tool.
00:07:36This is a built-in feature, so you can access it directly to perform a security review of all files. It follows its instructions and reports all findings.
00:07:44You can then prompt Claude to fix the identified issues, and it will implement the changes needed to make the app functional and secure.
00:07:51This is especially helpful for large-scale applications where security issues can be difficult to spot and might go unnoticed in production.
00:07:58With AI-generated code producing large volumes of changes, vulnerabilities could accidentally make it into production, and this tool helps prevent that.
00:08:06Also, if you are enjoying our content, consider pressing the hype button, because it helps us create more content like this and reach out to more people.
00:08:13The creator of Claude Code says that their team uses /commands for every inner-loop workflow that involves repetitive tasks.
00:08:21For example, they use a commit-pushpr/command that pushes code to the repo and opens a pull request.
00:08:26This command is open-source and available on the Claude Code plugin's official marketplace.
00:08:31You can access the commit command, which generates commits based on staged and unstaged changes, by following a specific series of steps.
00:08:38It contains the exact command the creator mentioned in his ex-post and includes the workflow for it as well.
00:08:44To install it, go to the plugin's marketplace, search for the commit command, and install it.
00:08:48Once installed, you can access all the commands available in the repo.
00:08:52When working with it, you can use these commands directly in your workflows to streamline repetitive tasks.
00:08:57In the workflow that the creator of Claude Code revealed, he actually mentioned that their team uses a tech debt skill.
00:09:04They advised running it at the end of every session to find duplicated code.
00:09:08This is not actually open-source and just used by the team internally.
00:09:12Since tech debt varies by the project it is being used in, the same went for them.
00:09:15But you can use it for your own purposes and tailor this skill to your own project.
00:09:19You can create your own for your own project, name it the same as he did,
00:09:23and add in your instructions for the skill using the Skill Creator skill, which is actually open-source.
00:09:28The instructions for the tech debt skill should contain a step-by-step workflow on how to identify changes,
00:09:33how it should actually detect duplicates, and how it should identify the tech debt that was collected during the session.
00:09:39It should also define how it handles the file structure relative to the project in detail,
00:09:43and it should end with verifications like running npm test and the linter.
00:09:47Now once all of this is in place, if you run the tech debt skill, it will go through your entire existing codebase,
00:09:52first using multiple agents to analyze it, and will report back the duplications in detail that are actually redundant and not needed.
00:09:59It then creates a shared library where all components can use the same feature and updates the components to access the same things from different places.
00:10:07Therefore, your session ends with a proper cleanup and a much more organized codebase with minimal tech debt.
00:10:12Now this skill that we created can be found in AI Labs Pro for this video and for all our previous videos,
00:10:18from where you can download and use it for your own projects.
00:10:21If you've found value in what we do and want to support the channel, this is the best way to do it.
00:10:26The link is in the description.
00:10:27Now the creators of Claude Code also use another skill called Ddupe.
00:10:30This skill is used internally by their team and is not available in any open source form or inside Claude Code.
00:10:36But from the results of the issue they linked in their tweet, we can reverse engineer what the skill possibly does,
00:10:42even though we cannot access the exact specialized system prompt.
00:10:45From what we observed, this skill checks whether a GitHub issue is a duplicate of an existing issue,
00:10:50and if it finds a duplicate, comments on it.
00:10:52It receives issues, parses the input, and searches for similar issues using the GitHub CLI.
00:10:57It contains criteria for determining what qualifies as similar or not.
00:11:01In case it doesn't wrongly qualify as an issue is duplicate or not,
00:11:04there is a set threshold which must be met before the issue is flagged as a duplicate.
00:11:08Once it finds a duplicate with around 70% certainty, it comments on that issue.
00:11:12The comment it makes on the duplicate issues includes information about why it matches and an explanation related to it.
00:11:18It still requires human verification though.
00:11:20So in case you are working on a project with multiple people
00:11:23and someone creates an issue that has already been dealt with,
00:11:26you can use this skill to handle the repeated ones,
00:11:29allowing you to focus on new issues instead of wasting time,
00:11:32clarifying that the issue has already been resolved.
00:11:34Now another skill that the team at Anthropic uses frequently is the Remotion skill inside Claude Code.
00:11:40This skill allows you to create motion graphics and videos with AI using simple prompts.
00:11:45You can install it and use it directly inside Claude Code, specifying how you want your animations to look.
00:11:50Now did you know that this skill is actually used across all product announcement videos
00:11:55and other marketing videos produced by the Claude marketing team?
00:11:58It's one of the most important skills in their workflow
00:12:00and it powers many of the team's non-development tasks as well.
00:12:04That brings us to the end of this video.
00:12:06If you'd like to support the channel and help us keep making videos like this,
00:12:09you can do so by using the super thanks button below.
00:12:12As always, thank you for watching and I'll see you in the next one.