Skip to main content

Getting visual explanations from your coding agent with the show-me skill

Coding agents are good at explaining things. They are also good at explaining them in five verbose paragraphs when a ten-line tree would have done the job. You would almost think that output tokens are free!

Recently I came across the show-me skill from HumanLayer. It’s a small skill with one purpose: when you ask your agent to explain something, it answers with a visual representation instead of a wall of text.

What does it do?

The whole skill is a short set of instructions. It tells the agent to skip the preamble, keep the prose brief and pick the smallest view that makes the point clear.

Sounds simple? Because the idea is indeed. For each kind of explanation, the skill names the view to use:

  • Logic or an algorithm → pseudocode
  • Runtime control flow → a call tree
  • UI structure → a component tree, including the state and module boundaries that matter
  • File responsibility or a broad refactor → a shallow file tree
  • Component interaction or data flow → a Mermaid diagram

Here is what that looks like in practice. A call tree for a form submit:

submitForm
  createSession
    persistPrompt
    launchAgent
  navigateToSession

And a shallow file tree to explain who owns what:

src/
├── commands/       # parses user actions
├── sessions/       # owns session state
└── transport/      # sends API requests

Nothing fancy. But compare that with the paragraph you would normally get back.

Installing the skill

The skill can be installed with the skills CLI:

npx skills add https://github.com/humanlayer/skills --skill show-me

After that you invoke it with a slash command like any other skill:

/show-me

Putting it into action

I (or one of my agents) did a big code change to add the authentication layer to my solution. This is a rather large changes involving multiple projects and files to be changed. Let's use /show-me to see what has changed:





Why I like this skill as an architect

As an architect I have to read a lot of code. Most of the time when I’m reviewing code changes, I’m not looking for an explanation. I’m looking for the shape of something: who calls whom, which module owns the state, where a refactor will touch. That is exactly what trees and sequence diagrams are good at, and prose is not.

The skill also forces the agent to be selective. You don’t get a diagram of everything, just the one that answers your question.

The skill also covers when to show a full code block instead of a fragment: when most of it is new, when leaving out context would hide ownership or order, or when you need a target shape you can copy. For larger explanations it can also produce a focused HTML page, which the agent opens for you.

Remark: The skill is plain instructions, nothing more. If you disagree with one of its choices (say you prefer ASCII over Mermaid), you can read the SKILL.md and adjust it for your own setup.

Should you use it?

If you regularly ask your agent to explain unfamiliar code, yes. It costs one command to install and it changes the format of every answer you ask for with /show-me.

More information

Labels:

Popular posts from this blog

Podman– Command execution failed with exit code 125

After updating WSL on one of the developer machines, Podman failed to work. When we took a look through Podman Desktop, we noticed that Podman had stopped running and returned the following error message: Error: Command execution failed with exit code 125 Here are the steps we tried to fix the issue: We started by running podman info to get some extra details on what could be wrong: >podman info OS: windows/amd64 provider: wsl version: 5.3.1 Cannot connect to Podman. Please verify your connection to the Linux system using `podman system connection list`, or try `podman machine init` and `podman machine start` to manage a new Linux VM Error: unable to connect to Podman socket: failed to connect: dial tcp 127.0.0.1:2655: connectex: No connection could be made because the target machine actively refused it. That makes sense as the podman VM was not running. Let’s check the VM: >podman machine list NAME         ...
Read more

Cache stampede: when our cache turned against us

While investigating some performance issues, we ran into an ASP.NET Core API that cached a fairly expensive aggregation query for 60 seconds. Under normal load, that was fine: one request rebuilds the cache, everyone else reads from it. Under peak load, dozens of requests would arrive in that same expiry window, all see a cache miss, and all fire the same expensive query in parallel. The database didn't like that. That was the moment when our caching layer stopped helping and started hurting. A burst of requests comes in at the same time, all miss the cache, and all go hammer the database or the downstream API at once. That's a cache stampede . The cache was supposed to protect our backend, and for a few hundred milliseconds it did the opposite. Why this happens IMemoryCache.GetOrCreate (and its async sibling) looks like it protects you, but it doesn't add any locking on its own. Look at the naive version: public async Task<Report> GetReportAsync(string key) ...
Read more

VS Code Planning mode

After the introduction of Plan mode in Visual Studio , it now also found its way into VS Code. Planning mode, or as I like to call it 'Hannibal mode', extends GitHub Copilot's Agent Mode capabilities to handle larger, multi-step coding tasks with a structured approach. Instead of jumping straight into code generation, Planning mode creates a detailed execution plan. If you want more details, have a look at my previous post . Putting plan mode into action VS Code takes a different approach compared to Visual Studio when using plan mode. Instead of a configuration setting that you can activate but have limited control over, planning is available as a separate chat mode/agent: I like this approach better than how Visual Studio does it as you have explicit control when plan mode is activated. Instead of immediately diving into execution, the plan agent creates a plan and asks some follow up questions: You can further edit the plan by clicking on ‘Open in Editor’: ...
Read more