This guide helps you create a client tool for an Atlas agent. A client tool runs in your browser and lets the agent fetch data, show charts, or update the screen. You do not need to write complex code because the skill handles the setup and wiring for you.
Before you start, make sure the integrate-atlas-chat step is done. That step puts all the needed files and packages into your project. Then you can define your tool using a schema from TypeBox. The schema tells the agent what information the tool needs.
Once your tool is defined, you add it to the useAtlasChat function. The agent then knows it can call your tool. The tool returns a text output and optional details that the app can show. This makes your agent more useful and interactive.
Global
mkdir -p ~/.claude/skills/create-client-toolProject
mkdir -p .claude/skills/create-client-toolSource Repository
Find Skillsvercel-labs/skills
Find and install the perfect skill to extend your AI agent
Microsoft Foundrymicrosoft/azure-skills
Build, deploy, and improve AI agents on Microsoft Foundry from start to finish
Azure Aimicrosoft/azure-skills
Search, transcribe, and analyze with Azure AI tools for smarter apps
Azure Hosted Copilot Sdkmicrosoft/azure-skills
Build, deploy, and manage your Copilot SDK apps on Azure with ease
Triagemattpocock/skills
Triage issues with a state machine driven by clear roles and agent briefs
Handoffmattpocock/skills
Hand off your work to another AI agent with a clear summary
Image Editagentspace-so/runcomfy-agent-skills
Smart router picks the best AI model for your image editing needs
Agentspaceagentspace-so/skills
See your AI agent's live folder from any browser instantly