secondsky/claude-skills / plugins/cloudflare-agents/skills/cloudflare-agents
cloudflare-agents
VerifiedBuild AI agents on Cloudflare Workers with MCP integration, tool use, and LLM providers.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 2.7 KB |
| references/advanced-features.md | markdown | 13.5 KB |
| references/agent-api.md | markdown | 2.9 KB |
| references/configuration-guide.md | markdown | 3.1 KB |
| references/error-catalog.md | markdown | 275 B |
| references/http-sse-guide.md | markdown | 2.0 KB |
| references/mcp-integration.md | markdown | 2.7 KB |
| references/patterns-concepts.md | markdown | 11.0 KB |
| references/state-management.md | markdown | 9.5 KB |
| references/websockets-guide.md | markdown | 3.0 KB |
| templates/basic-agent.ts | script | 383 B |
| templates/browser-agent.ts | script | 4.9 KB |
| templates/calling-agents-worker.ts | script | 3.7 KB |
| templates/chat-agent-streaming.ts | script | 1.5 KB |
| templates/hitl-agent.ts | script | 6.6 KB |
| templates/mcp-server-basic.ts | script | 4.0 KB |
| templates/rag-agent.ts | script | 4.8 KB |
| templates/react-useagent-client.tsx | other | 5.5 KB |
| templates/scheduled-agent.ts | script | 4.5 KB |
| templates/state-sync-agent.ts | script | 3.9 KB |
| templates/websocket-agent.ts | script | 4.1 KB |
| templates/workflow-agent.ts | script | 3.8 KB |
| templates/wrangler-agents-config.jsonc | other | 1.9 KB |
templates/calling-agents-worker.ts
View on GitHub// Worker that calls Agents using routeAgentRequest and getAgentByName
import { Agent, AgentNamespace, routeAgentRequest, getAgentByName } from 'agents';
interface Env {
MyAgent: AgentNamespace<MyAgent>;
ChatAgent: AgentNamespace<ChatAgent>;
}
// Worker fetch handler
export default {
async fetch(request: Request, env: Env): Promise<Response> {
const url = new URL(request.url);
// Pattern 1: Automatic routing with routeAgentRequest
// Routes to: /agents/:agent-name/:instance-name
// Example: /agents/my-agent/user-123
if (url.pathname.startsWith('/agents/')) {
const response = await routeAgentRequest(request, env);
if (response) {
return response;
}
return new Response("Agent not found", { status: 404 });
}
// Pattern 2: Custom routing with getAgentByName
// Example: /user/:userId/profile
if (url.pathname.startsWith('/user/')) {
const userId = url.pathname.split('/')[2];
// Authenticate first (CRITICAL)
const authHeader = request.headers.get('Authorization');
if (!authHeader) {
return new Response("Unauthorized", { status: 401 });
}
// Verify token and get authenticated userId
const authenticatedUserId = await verifyToken(authHeader);
if (!authenticatedUserId || authenticatedUserId !== userId) {
return new Response("Forbidden", { status: 403 });
}
// Get or create agent for this user
const agent = getAgentByName<Env, MyAgent>(env.MyAgent, `user-${userId}`);
// Pass request to agent
return (await agent).fetch(request);
}
// Pattern 3: Call agent methods directly (RPC)
if (url.pathname === '/api/process') {
const { userId, data } = await request.json();
const agent = getAgentByName<Env, MyAgent>(env.MyAgent, `user-${userId}`);
// Call custom method on agent
const result = await (await agent).processData(data);
return Response.json({ result });
}
// Pattern 4: Multi-agent communication
if (url.pathname === '/api/chat-with-context') {
const { userId, message } = await request.json();
// Get user agent
const userAgent = getAgentByName<Env, MyAgent>(env.MyAgent, `user-${userId}`);
// Get chat agent
const chatAgent = getAgentByName<Env, ChatAgent>(env.ChatAgent, `chat-${userId}`);
// User agent prepares context
const context = await (await userAgent).getContext();
// Chat agent generates response
const response = await (await chatAgent).chat(message, context);
return Response.json({ response });
}
return new Response("Not Found", { status: 404 });
}
} satisfies ExportedHandler<Env>;
// Helper: Verify JWT token
async function verifyToken(token: string): Promise<string | null> {
try {
// Implement your token verification logic
// Example: JWT verification, session validation, etc.
return "user-123"; // Return userId if valid
} catch (e) {
return null;
}
}
// Agent definitions
export class MyAgent extends Agent<Env> {
async onRequest(request: Request): Promise<Response> {
return Response.json({
agent: this.name,
message: "Hello from MyAgent"
});
}
async processData(data: any): Promise<any> {
return { processed: true, data };
}
async getContext(): Promise<any> {
return { user: this.name, preferences: this.state };
}
}
export class ChatAgent extends Agent<Env> {
async onRequest(request: Request): Promise<Response> {
return Response.json({ agent: this.name });
}
async chat(message: string, context: any): Promise<string> {
return `Response to: ${message} (with context: ${JSON.stringify(context)})`;
}
}
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |