agentsclimarketplace

Cli to skill

Skill mdonmez/skills/cli-to-skill

A curated list of skills designed to enhance AI Agents' capabilities across a range of instructions.

Install
npx -y skills add mdonmez/skills --skill cli-to-skill

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Converts any CLI tool into an opencode Agent Skill by exploring the command, documenting its subcommands and options, and generating a SKILL.md file. Use when the user says "turn <command> into a skill", "create a skill for <cli>", "skillify <tool>", or wants to wrap any CLI as a reusable skill.

SKILL.md

4.0 KB, 918 tokens by cl100k_base, as published. Nobody here has run it

CLI to Skill Generator

Generates a complete opencode Agent Skill (SKILL.md) for any CLI command, following the same format as existing skills (dub, resense, ffmpeg, youtube-transcript, etc.).

Workflow

1. Identify the command

The user provides a CLI tool name (e.g., gh, docker, kubectl, pnpm, cargo, uv).

If they just say the name, confirm which command they mean. If ambiguous (e.g., pip vs pip3), ask.

2. Check if the command is available

<command> --help
<command> --version

If not installed, search the web for installation instructions and note them in the skill.

3. Explore the CLI structure

Run help recursively to discover subcommands:

<command> --help
<command> <subcommand> --help    # for each subcommand

For deeply nested CLIs, explore at least 2 levels deep.

4. Research online (if needed)

If help output is incomplete, search the web for:

  • Official docs / man pages
  • Common usage patterns
  • Notable flags and options

5. Generate the SKILL.md

Create a new directory C:\Users\Donmez\.agents\skills\<name>\ with a SKILL.md file.

Use this template structure (modeled after the existing skills):

---
name: <name>
description: >
  <One-paragraph description covering what the CLI does, key subcommands,
   and explicit trigger terms.>
---

# <Title>

<1-2 sentence overview of the CLI tool.>

## Prerequisites

<Installation or system requirements.>

## Global Options

| Option | Alias | Description |
|--------|-------|-------------|

## Commands

Document every relevant subcommand with:
- Name and syntax
- Description of what it does
- All arguments and flags in a table
- At least one concrete example

### `<command> <subcommand>`

```bash
<command> <subcommand> [options]

Arguments:

ArgDescription

Options:

OptionAliasDescriptionDefault

Examples

<5-10 real, copy-pasteable examples covering the most common use cases.>

Quick Reference

TaskCommand

### 6. Follow established conventions

- **Frontmatter**: `name` matches the directory name. `description` is imperative and lists trigger keywords.
- **Tables**: Use pipe tables for options, args, and references.
- **Headers**: ATX headers (`##`, `###`).
- **Code blocks**: Language-tagged fenced code blocks.
- **Sections**: Frontmatter → Title → Prerequisites → Global Options → Commands → Examples → Quick Reference → (optional) Architecture/Development.

### 7. Place the skill

C:\Users\Donmez.agents\skills<name>\SKILL.md


## Gotchas

- **Don't document every single flag** — focus on the most useful ones. Omit rare or deprecated options.
- **Don't copy-paste raw `--help` output** into the skill. Synthesize it into readable documentation with explanations.
- **Include the install command** even if the tool is already installed (for reproducibility).
- **Prefer real examples** — run the command and include actual output or realistic scenarios.
- **For very large CLIs** (docker, kubectl, aws), focus on a subset of commands the user specifically asked about, and note the coverage scope in the skill description.
- **Update the system prompt's skill list** — not needed, opencode reads the directory automatically.

## Validation Checklist

- [ ] Frontmatter has `name` and `description`
- [ ] Description mentions specific CLI name and trigger keywords
- [ ] Installation/prerequisites section exists
- [ ] At least 3 commands documented with tables and examples
- [ ] At least 5 practical examples in quick reference
- [ ] No raw `--help` dumps — all content is human-readable
- [ ] File is at `C:\Users\Donmez\.agents\skills\<name>\SKILL.md`

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.