AI Skills

Reusable AI skills for agents like Claude Code and Cursor. Install them with a single command to enhance your AI workflows with structured, brand-aligned procedural knowledge.

What Are AI Skills?

Skills are Markdown instruction files that give AI agents domain-specific expertise. Each skill contains structured prompts, frameworks, and guidelines that an agent reads and applies when performing a task. Think of them as downloadable playbooks your AI can follow on demand.

Available Skills

Skill	Path	Description
Brand Guidelines	brand-guidelines/SKILL.md	The DigitalSpeed visual identity and styling system — logo, color, typography, photography, and graphic system. Use when generating any design asset, UI, layout, or code that must conform to the DigitalSpeed brand styling.
Brand Persona	brand-persona/SKILL.md	The official Digital Speed brand voice, values, and communication guidelines. Use as a foundation for any brand-aligned content.
Case Study Generator	case-study-generator/SKILL.md	Generates case studies using the Problem → Solution → Outcome framework, written in the Digital Speed voice.
Figma to Jira	figma-to-jira/SKILL.md	Orchestrates Jira ticket creation from Figma designs using Atomic Design principles. Use when initializing a project or syncing design specs to Jira for development.
License Checker	license-checker/SKILL.md	Audits node_modules for package licenses and produces a compliance report. Use when reviewing open-source dependencies for license risk.
Tutorial Generator	tutorial-generator/SKILL.md	Generates step-by-step tutorials and how-to guides in the Digital Speed voice. Use when asked to write a tutorial, setup guide, or walkthrough.

Installation

Via the Skills CLI

npx skills add digitalspeed/ai-skills

This downloads the skill files into your project .agents/skills/ directory, making them available to your AI agent via symlinks, e.g.

.claude
└── skills
    ├── brand-guidelines -> ../../.agents/skills/brand-guidelines
    └── brand-persona -> ../../.agents/skills/brand-persona
    ...

Usage

Combining Skills

Skills are designed to be composable. The Case Study Generator references the Brand Persona automatically, but you can explicitly combine any skills to layer capabilities:

Use @brand-persona to write a blog post about AI.

or

Apply the @case-study-generator skill using the @brand-persona persona to transform this raw text into a case study: [paste your content]

In practice, explicitly referencing is recommended because:

• It ensures the skill is actually loaded into the agent's context
• Automatic triggering depends on the agent's ability to discover and match skills, which varies by tool
• It removes ambiguity when you have multiple skills installed (e.g., brand-guidelines vs. brand-persona)

Example Using Cursor with the Claude Extension

1. Create a new project and install skills:

   mkdir blogs
   cd blogs
   npx skills add digitalspeed/ai-skills --skill brand-persona

2. Install the skill when prompted.

3. Verify the file exists: .cursor/skills/brand-persona/SKILL.md

4. Open the Claude extension panel in Cursor and try:

   Use @brand-persona to write a blog post about AI and write it to a markdown file.

The agent will find the correct persona and generate a markdown blog post about AI.

Example Using Figma to Jira

1. Create a project directory and install the skill:

   mkdir ai-skills-<project> && cd ai-skills-<project>
   npx skills add digitalspeed/ai-skills --skill figma-to-jira

2. Set up .mcp.json with both MCP servers:

   {
     "mcpServers": {
       "figma": {
         "type": "http",
         "url": "https://mcp.figma.com/mcp"
       },
       "jira": {
         "type": "stdio",
         "command": "uvx",
         "args": ["mcp-atlassian@latest"],
         "env": {
           "JIRA_URL": "https://digitalspeed.atlassian.net",
           "JIRA_USERNAME": "you@digitalspeed.co.uk",
           "JIRA_API_TOKEN": "your-api-token"
         }
       }
     }
   }

3. In Jira, create a new Scrum board: Create project → Software development → Scrum.

4. Set up CLAUDE.md with your project defaults:

   ## Project Config
   - Figma: <insert figma url of project here>
   - Jira project key: <insert jira project key here>

5. Authenticate the Figma MCP server via OAuth:

   1. Open Claude Code in your project directory.
   2. Type /mcp — you should see the Figma and Jira servers listed.
   3. Select figma → Authenticate — this opens the browser for OAuth.
   4. Once both servers show as connected, proceed to the next step.

6. Add the following to CLAUDE.md to pre-approve all tools the skill uses so it runs without confirmation prompts:

   ## Allowed Tools
   - mcp__figma__*
   - mcp__jira__*
   - Bash(python3:*)

7. Run Claude Code and invoke the skill:

   Use @figma-to-jira

   Claude will read your Figma design and create the corresponding Jira epics, tasks, and tickets.

How Skills Work

1. Discovery -- The AI agent reads the YAML frontmatter and instructions at the top of each skill file.
2. Triggering -- When your prompt matches a skill's domain (e.g., "write a case study"), the agent loads the relevant instructions.
3. Execution -- The agent follows the skill's framework, tone guidelines, and step-by-step workflow to produce the output.

Creating a New Skill

Each skill is a Markdown file with YAML frontmatter. Use this template:

---
name: your-skill-name
description: A concise summary of what the skill does and when to use it.
author: DigitalSpeed
---

# Skill Title

## Instructions
1. Define the step-by-step process the AI should follow.
2. Reference other skills by name if needed.
3. Specify tone, structure, and output format.

## Execution Workflow
- Step 1: Analyze input.
- Step 2: Apply the framework.
- Step 3: Refine output.
- Step 4: Return clean Markdown.

Updating Skills

After making changes to skill files upstream, pull the latest versions into any project that uses them:

npx skills update