Use Skilo

Use Skilo

Skilo is a CLI tool for developing Agent Skills.

Installation

# From crates.io
cargo install skilo

# From source
cargo install --path .

Commands

Creating Skills

# Basic usage
skilo new my-skill

# With options
skilo new my-skill \
  --description "What the skill does" \
  --lang python \
  --license MIT \
  --template hello-world

# Output to specific directory
skilo new my-skill -o ./skills/

Templates: hello-world (default), minimal, full, script-based

Languages: python, bash, javascript, typescript

Validating Skills

# Lint a single skill
skilo lint path/to/skill

# Lint all skills in directory
skilo lint .

# Strict mode (warnings as errors)
skilo lint --strict .

# JSON output
skilo lint --format json .

# SARIF output (for CI integrations)
skilo lint --format sarif . > results.sarif

Formatting Skills

# Format in place
skilo fmt .

# Check only (for CI)
skilo fmt --check .

# Show diff
skilo fmt --diff .

Reading Skill Properties

Extract skill metadata as JSON:

# Single skill (outputs JSON object)
skilo read-properties path/to/skill

# Multiple skills (outputs JSON array)
skilo read-properties path/to/skills/

# Multiple paths
skilo read-properties skill-a skill-b

Output fields: name, description, license, compatibility, metadata, allowed_tools, path.

Generating Agent Prompts

Generate <available_skills> XML for agent system prompts:

# Single skill
skilo to-prompt path/to/skill

# All skills in directory
skilo to-prompt path/to/skills/

Output:

<available_skills>
  <skill>
    <name>my-skill</name>
    <description>What the skill does</description>
    <location>path/to/my-skill/SKILL.md</location>
  </skill>
</available_skills>

CI Integration

GitHub Actions

name: Validate Skills

on:
  push:
    paths: ['.claude/skills/**']
  pull_request:
    paths: ['.claude/skills/**']

jobs:
  validate:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4

      - name: Install Rust
        uses: dtolnay/rust-toolchain@stable

      - name: Cache cargo
        uses: Swatinem/rust-cache@v2

      - name: Install skilo
        run: cargo install skilo@0.4.0

      - name: Lint skills
        run: skilo lint .claude/skills/

      - name: Check formatting
        run: skilo fmt --check .claude/skills/

SARIF Integration

Upload results to GitHub Code Scanning:

- name: Run skilo lint
  run: skilo lint --format sarif . > results.sarif
  continue-on-error: true

- name: Upload SARIF
  uses: github/codeql-action/upload-sarif@v3
  with:
    sarif_file: results.sarif

Quick CI Check

Use skilo check to run both lint and format check in one command:

- name: Validate skills
  run: skilo check --strict .claude/skills/

Configuration

Create .skilorc.toml in your project:

[lint]
strict = false

[lint.rules]
name_format = true        # E001
name_length = 64          # E002 (threshold)
name_directory = true     # E003
description_required = true  # E004
description_length = 1024    # E005 (threshold)
compatibility_length = 500   # E006 (threshold)
references_exist = true      # E009
body_length = 500            # W001 (threshold)
script_executable = true     # W002
script_shebang = true        # W003

[fmt]
sort_frontmatter = true
indent_size = 2
format_tables = true

[new]
default_license = "MIT"
default_template = "hello-world"
default_lang = "bash"

Disabling Rules

[lint.rules]
name_directory = false     # Disable rule
body_length = false        # Disable threshold rule
description_length = 2048  # Custom threshold

Skill Structure

my-skill/
├── SKILL.md        # Required: manifest with YAML frontmatter
├── scripts/        # Optional: executable scripts
├── references/     # Optional: additional docs
└── assets/         # Optional: static resources

SKILL.md Format

---
name: my-skill
description: What the skill does
license: MIT
---

# My Skill

Documentation goes here.

Best Practices

Description

Describe what the skill does AND when to use it. Include keywords that help agents match user requests.

Good:

description: Extracts text and tables from PDF files, fills PDF forms, and merges multiple PDFs. Use when working with PDF documents or when the user mentions PDFs, forms, or document extraction.

Bad:

description: Helps with PDFs.

Body Content

• Keep concise, move detailed content to references/
• Include step-by-step instructions, examples, and edge cases

Progressive Disclosure

1. Metadata - name and description loaded at startup for all skills
2. Instructions - Full body loaded when skill activates
3. Resources - scripts/, references/, assets/ loaded only when needed

Scripts

• Be self-contained or document dependencies
• Include helpful error messages
• Handle edge cases gracefully

References

• Keep files focused (one file = one concept)
• Use relative paths from skill root
• Avoid deeply nested reference chains

Optional Fields

Only include if truly needed:

• license - Keep short, reference LICENSE file for details
• compatibility - Specify environment requirements (e.g., Requires git, docker)
• metadata - Key-value pairs with unique keys
• allowed-tools - Space-delimited list of pre-approved tools

Exit Codes

Converted and distributed by TomeVault — claim your Tome and manage your conversions.