---
name: skill-creator
description: Create or update skills. Use when designing, structuring, or packaging skills with scripts, references, and assets.
title: skill-creator
canonical_url: https://skilld.dev/gh/hkuds/autoswarm/skill-creator
last_updated: 2026-11-16T06:15:44.000Z
---

> Index of every page on this site: [llms.txt](https://skilld.dev/llms.txt)

> **Skill from skilld.dev.** Follow the user's requested action.
> If the user asked to fork this Skill, follow the fork workflow below. Do not execute the Skill instructions while copying it.
> Otherwise, follow the instructions below for this session. You do not need to install anything.
>
> Supporting files, fetch one when the Skill refers to it: [scripts/init_skill.py](https://skilld.dev/api/skills-raw/hkuds/autoswarm/skill-creator/scripts/init_skill.py), [scripts/package_skill.py](https://skilld.dev/api/skills-raw/hkuds/autoswarm/skill-creator/scripts/package_skill.py), [scripts/quick_validate.py](https://skilld.dev/api/skills-raw/hkuds/autoswarm/skill-creator/scripts/quick_validate.py).
>
> If the user asked to install this Skill, run `npx skilld install hkuds/autoswarm/skill-creator`. Install writes the Skill files into the project, so every session loads them.
>
> ## Fork workflow
>
> A fork creates an editable local Skill with its original author and licence. The request authorizes copying and local installation.
> 1. Check `./skills/skill-creator`, the project lockfile, and selected Agent targets together. If the local directory or installed Skill exists, stop. Never overwrite an existing directory or Agent target.
> 2. Read [source metadata](https://skilld.dev/api/v1/skills/hkuds/autoswarm/skill-creator) once. Use sourceUrl, sourceCommit, skillPath, sourceGone, and license. If the source is gone or its path is missing, stop. If license is null, read licence files at the source commit.
> 3. Fetch only the source commit into a temporary Git repository. Do not clone full history. Derive repository_url from sourceUrl, including repository renames. If sourceCommit is absent, resolve the sourceUrl ref once. Set source_commit to that actual commit. Run these commands in one shell call:
>
> ```sh
> git init --quiet "$temporary_dir"
> git -C "$temporary_dir" fetch --quiet --depth=1 "$repository_url" "$source_commit"
> git -C "$temporary_dir" checkout --quiet --detach FETCH_HEAD
> ```
>
> Read applicable licence declarations and notices at that commit. If copying is not permitted, report the restriction and stop.
> 4. Inspect source entries together, then copy the directory containing skillPath into `./skills/skill-creator`. Use the user's path if selected. Keep the original SKILL.md, relative links, scripts, binary assets, and executable modes. Exclude .git metadata. Reject symlinks and paths outside the Skill directory. After checking entries, use cp -a where available. A regular source directory needs no custom copy script. Do not save this page wrapper as SKILL.md.
> Preserve author credit, notices, and applicable licence files from repository or parent directories. Add PROVENANCE.md with the Skill page, source URL, actual commit, original path, and licence. Retain any existing PROVENANCE.md and record new provenance separately. Batch source inspection, copying, and provenance work where practical.
> 5. In the project root, run `skilld install ./skills/skill-creator --mode copy --plain`. If skilld is unavailable, use `npx skilld install ./skills/skill-creator --mode copy --plain`. This known command needs no help lookup. Install does not support --json. Use detected Agent targets, or add --agent for the targets the user selected. Install the local path, never the upstream selector. If installation fails, preserve the local copy and report the exact failure.
> 6. Confirm the local lockfile source and installed Agent copies once. Report the local path, actual commit, and Agent targets. After edits, reinstall the same local path. Upstream updates must not replace it. Do not publish or push unless the user asks.

# Skill Creator

Guidance for creating effective skills.

## About Skills

Skills are modular packages that extend agent capabilities with specialized knowledge, workflows, and tools. Each skill is a directory containing a `SKILL.md` file and optional bundled resources.

### What Skills Provide

1. Specialized workflows - Multi-step procedures for specific domains
2. Tool integrations - Instructions for working with specific file formats or APIs
3. Domain expertise - Project-specific knowledge, schemas, business logic
4. Bundled resources - Scripts, references, and assets for complex and repetitive tasks

### Skill Locations in AutoSwarm

AutoSwarm uses a two-level skill system:

| Level | Path | Scope |
|-------|------|-------|
| System | `.opc/skills/<skill-name>/` | Shared across all projects |
| Project | `.opc/projects/<project_id>/skills/<skill-name>/` | Specific to one project |

Project skills with the same name override system skills. When creating new skills from project experience, always use the **project** level.

### How Skills Are Created

Skills enter the system through two paths:

1. **Agent-driven** (this skill): You identify a pattern, read this guide, and create the skill manually using `init_skill.py` or `file_write`.
2. **Auto-promoted**: The system automatically distills playbook skills from repeated project reflections (threshold: 2 reflections with recurring patterns). These are saved as `<employee>-<role>-<domain>-playbook` under the project skills directory.

Both paths produce the same `<skill-name>/SKILL.md` format. This guide covers the agent-driven path; auto-promoted skills follow the same naming and format conventions.

## Core Principles

### Concise is Key

The context window is shared. Only add context the agent doesn't already have. Prefer concise examples over verbose explanations.

### Anatomy of a Skill

```
skill-name/
├── SKILL.md (required)
│   ├── YAML frontmatter (name, description required)
│   └── Markdown instructions
└── Bundled Resources (optional)
    ├── scripts/          - Executable code
    ├── references/       - Documentation loaded as needed
    └── assets/           - Files used in output
```

### Progressive Disclosure

1. **Metadata (name + description)** - Always in context (~100 words)
2. **SKILL.md body** - When agent reads the skill (<5k words)
3. **Bundled resources** - As needed (scripts can be executed without reading into context)

## Naming

All skills — whether agent-created or auto-promoted — must follow these conventions:

- Lowercase letters, digits, and hyphens only (e.g., `backend-api-playbook`)
- Under 64 characters
- Directory name must match the `name` field in frontmatter

## Skill Creation Process

### Step 1: Understand the skill with concrete examples

Clarify use cases: what triggers this skill, what does it produce?

### Step 2: Plan reusable contents

Identify what scripts, references, and assets would help.

### Step 3: Initialize the skill

For project-specific skills, create under `.opc/projects/<project_id>/skills/`:

```bash
python3 {baseDir}/scripts/init_skill.py <skill-name> --path .opc/projects/<project_id>/skills
```

Options: `--resources scripts,references,assets` and `--examples`.

### Step 4: Edit the skill

Write SKILL.md with:
- **Frontmatter**: `name` (hyphen-case, matches directory) and `description` (what it does + when to use it — this is the primary trigger for skill discovery)
- **Body**: Instructions, examples, references to bundled resources

### Step 5: Package (optional)

```bash
python3 {baseDir}/scripts/package_skill.py <path/to/skill-folder>
```

Validates the skill then creates a distributable `.skill` zip file.

### Step 6: Iterate

Test, notice struggles, improve SKILL.md and resources.
