Nushell logo

Nushell

OrganizationPopular
ccusage
nushell

Guides ccusage Nushell scripts. Use when adding, editing, formatting, or validating .nu files under .github/scripts, apps/ccusage/scripts, or scripts, including their Nix shebangs and GitHub Actions callers.

Overview

Publisherccusage
Repositoryccusage
Skill namenushell
Stars
19K
Forks
871
Bundled files
Instructions only
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by ccusage on GitHub. Read the source before you install it.

Installation

Install the Nushell AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/ccusage/ccusage/tree/main/.agents/skills/nushell
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/ccusage/ccusage.git /tmp/ccusage
mkdir -p .claude/skills
cp -r /tmp/ccusage/.agents/skills/nushell .claude/skills/nushell
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Nushell in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Nushell on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Nushell is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

ccusage Nushell

Nushell is a functional, structured-data language that happens to be a shell. Write it that way: pipelines of transformations over records and tables, not bash with different syntax. Root AGENTS.md says when to pick Nushell over Babashka, Rust, or TypeScript.

Scripts live in .github/scripts/, apps/ccusage/scripts/, and scripts/.

Runtime Shape

Every executable script is a .nu file with a Nix shebang and a def main; that shebang is how CI gets an interpreter, rather than a nushell profile install in the workflow. List only the tools the script directly invokes; --inputs-from is relative to the script's own directory, so its depth differs per directory — copy the header from a neighbor, such as .github/scripts/upsert-pr-comment.nu or apps/ccusage/scripts/stage-native-package.nu.

Files that exist only to be imported as modules carry no shebang and no main: .github/scripts/pricing-lock.nu, apps/ccusage/scripts/native-binary.nu.

Workflows run these scripts through the shebang, but the commit-msg hook in nix/git-hooks.nix invokes scripts/validate-commit-scope.nu as an argument to nushell instead, so that script only ever gets nu itself.

Calling external tools such as gh, jq, git, hyperfine, pnpm, node, or bun is fine when they are the right boundary — pin them through the shebang shell rather than a global install.

Docs

Read before writing. The sidebar on any of these reaches the rest of the book, plus /commands/ and /cookbook/.

https://www.nushell.sh/book/thinking_in_nu.html

https://www.nushell.sh/book/nushell_map_functional.html

https://www.nushell.sh/book/style_guide.html

Style

Functional style is the default. Treat each of these as a defect to fix:

  • mut + for as an accumulator. Use reduce, each, where, group-by, zip, flatten, insert/update, generate. mut is legitimate only for genuinely sequential state that no filter expresses.
  • String plumbing between steps. Pass records and tables; serialize once at the boundary (to json --raw) and parse once on the way in (from json).
  • Nested if/else chains on a value's shape. Use match, including list patterns and guards.
  • Shelling out for data. Native commands over ^jq, ^sed, ^awk, ^date.
  • Bare each for effects. Say so with | ignore when the result is unused.

Also:

  • Type custom command signatures — parameter types, --flag, and the return type. They document intent and are checked at parse time.
  • run-external for commands whose flags Nushell would otherwise parse; quote short flags such as '-L', '-x', '-c', and keep arguments as lists up to the external boundary.
  • complete when you need exit code, stdout, and stderr without throwing; error make rather than a sentinel return value.
  • Timestamps are datetime and durations are duration; compare and subtract them directly instead of formatting to strings.
  • Progress to stderr for CI scripts whose stdout is a data artifact.

Validation

Nushell renames and breaks things between minor releases, so check constructs against the pinned interpreter instead of trusting memory:

sh
direnv exec . nu --ide-check 10 path/to/script.nu
direnv exec . nu -c 'help <command>'

just fmt runs nufmt over *.nu through treefmt. For behavior changes, invoke the script through its shebang as a smoke check and run the repo check that owns the caller.

Frequently asked questions

What does the Nushell AI skill do?

Guides ccusage Nushell scripts. Use when adding, editing, formatting, or validating .nu files under .github/scripts, apps/ccusage/scripts, or scripts, including their Nix shebangs and GitHub Actions callers.

Why use Nushell on TypingMind?

Because you install it once and use it with any model. Nushell is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Nushell in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/ccusage/ccusage/tree/main/.agents/skills/nushell. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Nushell?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Nushell?

As many as you like. As long as a model supports skills, you can use Nushell with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Nushell AI skill free?

It is published on GitHub by ccusage. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇