@useatlas/create 0.0.1

package/templates/nextjs-standalone/src/ui/lib/types.ts

@@ -0,0 +1,145 @@
+/**
+ * Shared types for Atlas UI.
+ *
+ * These are duplicated from @atlas/api to keep the frontend package independent
+ * of the API package. The API is the source of truth — keep these in sync.
+ */
+
+// --- Auth types (from @atlas/api/lib/auth/types) ---
+
+export const AUTH_MODES = ["none", "simple-key", "managed", "byot"] as const;
+export type AuthMode = (typeof AUTH_MODES)[number];
+
+// --- Conversation types (from @atlas/api/lib/conversation-types) ---
+
+export type MessageRole = "user" | "assistant" | "system" | "tool";
+export type Surface = "web" | "api" | "mcp" | "slack";
+
+export interface Conversation {
+  id: string;
+  userId: string | null;
+  title: string | null;
+  surface: Surface;
+  connectionId: string | null;
+  starred: boolean;
+  createdAt: string;
+  updatedAt: string;
+}
+
+export interface Message {
+  id: string;
+  conversationId: string;
+  role: MessageRole;
+  content: unknown;
+  createdAt: string;
+}
+
+export interface ConversationWithMessages extends Conversation {
+  messages: Message[];
+}
+
+// --- Error types (from @atlas/api/lib/errors) ---
+
+export const CHAT_ERROR_CODES = [
+  "auth_error",
+  "rate_limited",
+  "configuration_error",
+  "no_datasource",
+  "invalid_request",
+  "provider_model_not_found",
+  "provider_auth_error",
+  "provider_rate_limit",
+  "provider_timeout",
+  "provider_unreachable",
+  "provider_error",
+  "internal_error",
+] as const;
+
+export type ChatErrorCode = (typeof CHAT_ERROR_CODES)[number];
+
+export interface ChatErrorInfo {
+  title: string;
+  detail?: string;
+  retryAfterSeconds?: number;
+  code?: ChatErrorCode;
+}
+
+function authErrorMessage(authMode: AuthMode): string {
+  switch (authMode) {
+    case "simple-key":
+      return "Invalid or missing API key. Check your key and try again.";
+    case "managed":
+      return "Your session has expired. Please sign in again.";
+    case "byot":
+      return "Authentication failed. Your token may have expired.";
+    case "none":
+      return "An unexpected authentication error occurred. Please refresh the page.";
+    default: {
+      const _exhaustive: never = authMode;
+      return `Authentication failed (unknown mode: ${_exhaustive}).`;
+    }
+  }
+}
+
+export function parseChatError(error: Error, authMode: AuthMode): ChatErrorInfo {
+  let parsed: Record<string, unknown>;
+  try {
+    parsed = JSON.parse(error.message);
+  } catch {
+    return { title: "Something went wrong. Please try again." };
+  }
+
+  const code = typeof parsed.error === "string" ? parsed.error : undefined;
+  const serverMessage = typeof parsed.message === "string" ? parsed.message : undefined;
+
+  switch (code) {
+    case "auth_error":
+      return { title: authErrorMessage(authMode), code };
+
+    case "rate_limited": {
+      const raw = typeof parsed.retryAfterSeconds === "number" ? parsed.retryAfterSeconds : undefined;
+      const clamped = raw !== undefined ? Math.max(0, Math.min(raw, 300)) : undefined;
+      return {
+        title: "Too many requests.",
+        detail: clamped !== undefined
+          ? `Try again in ${clamped} seconds.`
+          : "Please wait before trying again.",
+        retryAfterSeconds: clamped,
+        code,
+      };
+    }
+
+    case "configuration_error":
+      return { title: "Atlas is not fully configured.", detail: serverMessage, code };
+
+    case "no_datasource":
+      return { title: "No data source configured.", detail: serverMessage, code };
+
+    case "invalid_request":
+      return { title: "Invalid request.", detail: serverMessage, code };
+
+    case "provider_model_not_found":
+      return { title: "The configured AI model was not found.", detail: serverMessage, code };
+
+    case "provider_auth_error":
+      return { title: "The AI provider could not authenticate.", detail: serverMessage, code };
+
+    case "provider_rate_limit":
+      return { title: "The AI provider is rate limiting requests.", detail: serverMessage, code };
+
+    case "provider_timeout":
+      return { title: "The AI provider timed out.", detail: serverMessage, code };
+
+    case "provider_unreachable":
+      return { title: "Could not reach the AI provider.", detail: serverMessage, code };
+
+    case "provider_error":
+      return { title: "The AI provider returned an error.", detail: serverMessage, code };
+
+    case "internal_error":
+      return { title: serverMessage ?? "An unexpected error occurred.", code };
+
+    default:
+      return { title: serverMessage ?? "Something went wrong. Please try again." };
+  }
+}

package/templates/nextjs-standalone/tsconfig.json

@@ -0,0 +1,32 @@
+{
+  "compilerOptions": {
+    "target": "ES2022",
+    "lib": ["dom", "dom.iterable", "esnext"],
+    "allowJs": true,
+    "skipLibCheck": true,
+    "strict": true,
+    "noEmit": true,
+    "esModuleInterop": true,
+    "module": "esnext",
+    "moduleResolution": "bundler",
+    "resolveJsonModule": true,
+    "isolatedModules": true,
+    "jsx": "react-jsx",
+    "incremental": true,
+    "plugins": [{ "name": "next" }],
+    "paths": {
+      "@/*": ["./src/*"],
+      "@atlas/api/app": ["./src/api/index.ts"],
+      "@atlas/api/*": ["./src/*"]
+    }
+  },
+  "include": [
+    "next-env.d.ts",
+    "src/**/*.ts",
+    "src/**/*.tsx",
+    "bin/**/*.ts",
+    ".next/types/**/*.ts",
+    ".next/dev/types/**/*.ts"
+  ],
+  "exclude": ["node_modules"]
+}

package/templates/nextjs-standalone/vercel.json

@@ -0,0 +1,4 @@
+{
+  "framework": "nextjs",
+  "crons": [{ "path": "/api/v1/scheduled-tasks/tick", "schedule": "* * * * *" }]
+}