diff --git a/docs.json b/docs.json index 1305a972..2e87efe3 100644 --- a/docs.json +++ b/docs.json @@ -42,6 +42,7 @@ "get-started/concepts", "get-started/manage-accounts", "get-started/api-keys", + "get-started/agent-skills", "get-started/mcp-servers" ] }, diff --git a/get-started/agent-skills.mdx b/get-started/agent-skills.mdx new file mode 100644 index 00000000..931a391d --- /dev/null +++ b/get-started/agent-skills.mdx @@ -0,0 +1,86 @@ +--- +title: "Agent skills for AI coding tools" +sidebarTitle: "Agent skills" +description: "Manage GPU workloads on Runpod with coding agents like Claude Code and Cursor." +tag: "NEW" +--- + +Agent skills make it easier for coding agents to use Runpod directly from your editor or terminal. Once installed, you can ask your agent to create Pods, deploy Serverless endpoints, manage templates, and more, all through natural language. + +These skills work with [Claude Code](https://docs.anthropic.com/en/docs/claude-code), [Cursor](https://www.cursor.com/), [GitHub Copilot](https://github.com/features/copilot), [Windsurf](https://codeium.com/windsurf), [Cline](https://github.com/cline/cline), and [17+ other AI agents](https://skills.sh/). + +## Installation + +Install the Runpod skills package in your project: + +```bash +npx skills add runpod/skills +``` + +Restart your agent after installation. + +If you plan to use the `runpodctl` skill, you also need to [install the Runpod CLI](/runpodctl/overview) and run the setup command to configure your [API key](/get-started/api-keys) and SSH keys: + +```bash +runpodctl doctor +``` + +## Available skills + +The Runpod skills package includes two skills: + +| Skill | Description | +|-------|-------------| +| **Flash** | Build and deploy AI workloads using the `runpod-flash` SDK. Write code locally and deploy to remote GPUs/CPUs. | +| **runpodctl** | Manage GPU Pods, Serverless endpoints, templates, network volumes, and models through the Runpod CLI. | + +## What you can do + +Once installed, you can ask your AI agent to perform tasks like the following: + +| Category | Example prompt | +|----------|----------------| +| Create resources | "Create a Pod with an RTX 4090" | +| List resources | "List my Pods" or "Show my Serverless endpoints" | +| GPU availability | "What GPUs are available?" | +| Account info | "Show my account balance" | +| Deploy endpoints | "Deploy a Serverless endpoint using my template" | +| Manage Pods | "Stop my Pod" or "SSH into my Pod" | +| Deploy with Flash | "Deploy this function to a remote GPU with Flash" | +| Local development | "Start a local dev server with Flash" | + +### Flash skill + +The Flash skill lets your agent use the [runpod-flash](/flash/overview) SDK to build and deploy functions to Runpod Serverless. Your agent can generate and deploy code like the following: + +```python +from runpod_flash import Endpoint, GpuGroup + +@Endpoint(name="my-worker", gpu=GpuGroup.AMPERE_80, workers=5, dependencies=["torch"]) +async def compute(data): + import torch + return {"sum": torch.tensor(data, device="cuda").sum().item()} +``` + +For more details on endpoint types and configuration, see [Create endpoints with Flash](/flash/create-endpoints). + +### runpodctl skill + +The runpodctl skill gives your agent access to the full [Runpod CLI](/runpodctl/overview) for managing Pods, Serverless endpoints, templates, network volumes, and file transfers. + +## Learn more + + + + Source code and full skill definitions. + + + The skills platform with the full list of compatible AI agents. + + + Full runpodctl documentation. + + + Another way to integrate AI tools with Runpod. + + diff --git a/get-started/mcp-servers.mdx b/get-started/mcp-servers.mdx index 9f62e7e8..a9bc7da5 100644 --- a/get-started/mcp-servers.mdx +++ b/get-started/mcp-servers.mdx @@ -1,6 +1,6 @@ --- title: "Use Runpod's MCP servers" -sidebarTitle: "Runpod MCP servers" +sidebarTitle: "MCP servers" description: "Connect AI tools to Runpod using the Model Context Protocol for infrastructure management and documentation access." tag: "NEW" ---