Skip to content
Back to skills

Source Command Generate Llms

ASecurity

Regenerate llms.txt and web/public/llms.txt from current README and docs/llms-full.txt — keeps AI-discovery files in sync with the product

  • 6 stars
  • 0 votes
  • 0 copies
  • 3 views
  • Added September 2, 2026
ai-agentsgogit

Security analysis

A100/100

Scanned September 2, 2026

npx -y skills add opencue/cuecards --skill source-command-generate-llms --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Source Command Generate Llms?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Source Command Generate Llms
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/opencue-source-command-generate-llms/badge)](https://www.skillsdirectory.com/skills/opencue-source-command-generate-llms)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: "source-command-generate-llms"
description: "Regenerate llms.txt and web/public/llms.txt from current README and docs/llms-full.txt — keeps AI-discovery files in sync with the product"
---

# source-command-generate-llms

Use this skill when the user asks to run the migrated source command `generate-llms`.

## Command Template

# /generate-llms — Regenerate llms.txt Files

Regenerate `llms.txt` (root) and `web/public/llms.txt` from current source material.
Run this after any significant change to the README, profiles list, or agent support.

## Source files to read first

1. `README.md` — product description, tagline, install command, agent list
2. `docs/llms-full.txt` — dense technical summary (authoritative)
3. `profiles/` — list current profiles with `ls profiles/ | grep -v '^_'`
4. `package.json` — current version and npm package name

## Root llms.txt — what it must contain

```
# cuecards (cue)

<one-paragraph description — from README tagline + what-it-does>

## What it solves
<2–3 sentences on the problem>

## Core docs
- [README](<url>) — <purpose>
- [llms-full.txt](<url>) — <purpose>
- [AGENTS.md](<url>) — <purpose>
- [launch.md](<url>) — <purpose>

## Setup guides
- [macOS](<url>)
- [Linux / WSL2](<url>)
- [Windows](<url>)

## Comparisons
- [cue vs Codex-switcher](<url>)
- [cue vs skillport](<url>)
- [cue vs Kiro Powers](<url>)

## Repository
- GitHub: https://github.com/opencue/cuecards
- npm: https://www.npmjs.com/package/cue-ai
- License: MIT
```

## web/public/llms.txt — condensed version

Same structure as root `llms.txt` but:
- Shorter descriptions (1 sentence max per link)
- Point to deployed URLs (`https://opencue.github.io/cuecards/...`) where they exist
- Include the `## GEO files` section pointing to `content/geo/`

## After generating

1. Diff the new files against the previous versions: `git diff llms.txt web/public/llms.txt`
2. Verify no URLs were broken (spot-check 3 links)
3. Commit as: `docs: refresh llms.txt [skip ci]`

## Automation note

This command is also triggered by the nightly `discover` workflow. If running
manually, check the workflow hasn't already run today:
`git log --oneline --since="24 hours ago" | grep "llms"`

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.