Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.replit.com/llms.txt

Use this file to discover all available pages before exploring further.

Skills teach Agent new capabilities — preferred patterns, library specifics, design system rules, or solutions to problems you’ve already solved. Attach one to a message, install one in your project, or pick one as a starting point for a new project.

Use a skill in chat

The fastest way to try skills is to attach one to a single message — no installation needed.
Skill picker opened from the + button next to the chat input, showing the Skills list with Ad Creative, AI Recruiter, AI SDR, and AI Secretary
1

Open the skill picker

Click the + button next to the chat input and select Use a skill.
2

Browse and select a skill

Browse skills by category — Business & finance, Creative, Research & analysis, and more. Select one or more skills to attach to your message.
3

Send your message

Type your request and send. Agent uses the selected skills for that message only. A skill chip appears on your message so you can see which skills are active.
For example, attach Stock Analyzer and ask Agent to build an equity research report, or attach Deep Research to have Agent investigate a topic with structured findings.
Skills attached in chat don’t install anything in your project. They’re available instantly in every project.

Start a project with a skill

When you create a new project, you can pick a pre-vetted skill as the starting point. Agent uses the skill’s instructions to scaffold the project.
1

Open the starting point picker

On the new-project screen, click + and select Use a skill.
2

Pick a skill

Browse curated skills like AI SDR, SEO Auditor, Ad Creative, or Programmatic SEO. Select the one that matches the kind of app you want to build.
3

Describe your app

Tell Agent what you want. Agent builds the project with the skill’s patterns and conventions applied from the start.

Install a skill in your project

When you want Agent to apply a skill consistently across every conversation in a project, install it. Installed skills live in your project’s /.agents/skills directory and persist across chat sessions.

Install from the Skills pane

Open the Skills pane in the Project Editor and select the Discover tab to search community-contributed skills. Find one you want, click Install, and it’s added to your project automatically.
Discover skills tab showing searchable community-contributed skills with install buttons
For example, when building a portfolio site with advanced animations, you might install:
  • GSAP React — teaches Agent scroll-triggered animations, text reveals, and SVG path drawing
  • Tailwind design system — gives Agent knowledge of Tailwind CSS patterns for typography, spacing, and view transitions
  • Find skills — teaches Agent how to discover and install new skills on your behalf
You can also install skills via the npx skills CLI:
npx skills <skill> -a replit

Create a skill from a conversation

The most natural way to create a skill is through conversation with Agent. After solving a problem together or researching a new library, ask Agent to capture what it learned: Agent uses the full conversation context to write a detailed skill file. This works particularly well after debugging sessions where you’ve built up shared understanding of a problem.

Write a custom skill

For advanced use cases, write skills directly following the Agent Skills specification. Toggle Show Hidden Files in the file sidebar, open /.agents/skills/, and create a new Markdown file. This gives you complete control over what Agent knows and how it behaves.

Next steps

Agent skills (concepts)

Learn when to use skills, proactive vs reactive patterns, and how skills compare to MCP servers.

Agent Skills reference

Technical details on skill structure, management, and how skills work under the hood.

Browse community skills

Find skills built by the community at skills.sh.

Agent Skills specification

Read the open standard so you can write skills that work across any agent.