austere-setup
Skill8 starsChanged 4 days agoMITView on GitHub
Set up austere-axi on this machine for the user. Checks Node, installs the `austere` skill, optionally installs the CLI globally, then proves the review loop works with a test page. Use when the user asks to install, set up or configure austere-axi.
- Installs packages
In this repository5 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- add-componentSkill
- review-changeSkill
- austereSkill
The file
We split this skill into the rules it gives an agent
10 rules. About 0 are common, 4 are distinctive, and 6 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 11
Ask the user only the questions in steps 3 and 4.
- Distinctive: line 19
If Node is older than 22.18, stop.
- Very specific: line 23
Run npx -y austere-axi --help.
- Very specific: line 29
If the austere skill is already available (for example from the austere Claude Code plugin), skip this step.
- Distinctive: line 31
Ask: install it for all projects (global), or only this project?
- Very specific: line 38
If the skills tool asks which agents to install for, pick the agent the user works with.
- Distinctive: line 42
Ask the user about each item.
- Very specific: line 45
Phone review: If Tailscale is running, the server also listens on the tailnet, and open prints a phone link.
- Very specific: line 51
Write austere-setup-test.html in the current directory:
- Very specific: line 72
If the user cannot open the page, read the server log in ~/.austere-axi/ and report it.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
austere-setup
Set up austere-axi for the user. Do the steps in order. Stop and tell the user when a step fails.
Ask the user only the questions in steps 3 and 4. Use the default for each if they do not answer.
1. Check Node
Run node --version.
- Node 22.18 or later is necessary.
If Node is older than 22.18, stop. Tell the user to install a current Node LTS.
2. Check the CLI
Run npx -y austere-axi --help. It must print the WORKFLOW section. If npm cannot reach the registry, stop and show the error.
3. Install the austere skill
This skill tells agents when to make a page, and adds the /austere command.
If the austere skill is already available (for example from the austere Claude Code plugin), skip this step.
Ask: install it for all projects (global), or only this project? Default: global.
npx skills add MandalorianBatman/austere-axi --skill austere -g # global
npx skills add MandalorianBatman/austere-axi --skill austere # this projectIf the skills tool asks which agents to install for, pick the agent the user works with.
4. Optional choices
Ask the user about each item. Default: no.
- Global command.
npm install -g austere-axi. Thenaustere-axiworks withoutnpx -y. Update it later withnpm update -g austere-axi. - Phone review. If Tailscale is running, the server also listens on the tailnet, and
openprints a phone link. Nothing to install. Runtailscale statusto check.
On a machine with no screen (SSH, CI, container), tell the user to set AUSTERE_AXI_NO_OPEN=1, so the CLI prints links and does not try to open a browser. A browser the CLI did not open gets "403" at a printed link until it signs in: run npx -y austere-axi login <file> and give the user the link it prints. It works once, within 60 seconds.
5. Prove it works
Write austere-setup-test.html in the current directory:
<!doctype html>
<html><head><meta charset="utf-8"><title>austere-axi works</title>
<script src="https://cdn.jsdelivr.net/npm/austere-axi@0/dist/austere.js"></script></head>
<body><script type="text/openui">
root = Stack([card], "column", "l")
card = Card([CardHeader("austere-axi works", "Setup test"), text, buttons])
text = MarkDownRenderer("Click a part of this page to annotate it, or press the button.")
buttons = Buttons([Button("It works")])
</script></body></html>Then:
npx -y austere-axi check austere-setup-test.htmlmust report no errors.npx -y austere-axi austere-setup-test.htmlopens the review. Give the user the printed link. If it did not open a browser (browser: not opened), runnpx -y austere-axi login austere-setup-test.htmland give the user that link instead.npx -y austere-axi poll austere-setup-test.htmlwaits for the user. Ask them to press It works. The poll returns an item for the button.npx -y austere-axi end austere-setup-test.html, then delete the file.
If the user cannot open the page, read the server log in ~/.austere-axi/ and report it.
6. Report
Tell the user, in a short list:
- the Node version
- where the
austereskill is installed - which optional items you set up
- that the test page worked
To make a page later, the user asks for one, or types /austere <what to show>.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

