@hasura/promptql 2.0.0-alpha.40 → 2.0.0-alpha.42

package/README.md

@@ -1,21 +1,10 @@
 # PromptQL TypeScript SDK
 
-The PromptQL TypeScript SDK is a general-purpose SDK that interacts with the [PromptQL API](https://promptql.hasura.io) by Hasura. PromptQL enables natural language querying of your data through an AI-powered interface.
+The PromptQL TypeScript SDK is a general-purpose SDK that interacts with the [PromptQL API](https://promptql.io) by Hasura. PromptQL is a multiplayer AI agent — think Claude or ChatGPT, but with shared threads and a shared brain.
 
 > [!NOTE]
 > This SDK supports PromptQL v2. Please install [@hasura/promptql@1.0.0](https://www.npmjs.com/package/@hasura/promptql/v/1.0.0) if you wants to integrate PromptQL v1.
 
-## Features
-
-- **Type-safe API Client**: Fully typed TypeScript SDK with auto-generated types from GraphQL schema
-- **Thread Management**: Create, list, and manage conversation threads
-- **Real-time Streaming**: Subscribe to thread events with GraphQL subscriptions using RxJS observables
-- **Authentication**: Built-in authentication handling with service account tokens
-- **Query Planning**: Access to AI agent query plans and execution steps
-- **Artifact Support**: Handle modified artifacts from AI responses
-- **CLI Tool**: Command-line interface for quick interactions
-- **Flexible Configuration**: Support for custom base URLs, headers, and fetch implementations
-
 ## Requirements
 
 - Node.js >= 24
@@ -42,16 +31,14 @@ bun add @hasura/promptql@latest
 
 ## Quick Start
 
-### Getting Service Account Token
+### Getting PromptQL access token
 
-1. Visit your Hasura project console
-2. Navigate to Project Settings
-3. Go to Service Accounts section
-4. Create a new service account token with `Read Only` permission. Avoid using the `Admin` role. Hackers can modify your project metadata if the service account token is leaked.
+1. Visit your PromptQL project
+2. Navigate to User Directory
+3. Use your account or create a bot account.
+4. Create a personal access token with `MCP` scope.
 5. Copy the token and use it in your SDK configuration
 
-For more details, check the [Hasura documentation](https://hasura.io/docs/3.0/project-configuration/project-management/service-accounts/#how-to-create-service-account).
-
 ### Basic Usage
 
 #### Start a PromptQL thread
@@ -61,7 +48,8 @@ import { PromptQLSdk } from '@hasura/promptql';
 
 // Initialize the SDK
 const sdk = new PromptQLSdk({
-  serviceAccountToken: 'your-service-account-token',
+  baseUrl: 'https://your-data-plane-url',
+  accessToken: 'personal-access-token',
 });
 
 // Start a new thread
@@ -109,7 +97,6 @@ import { lastValueFrom } from "rxjs";
 
 // Initialize the SDK
 const sdk = new PromptQLSdk({
-  projectId: 'your-project-id',
   serviceAccountToken: 'your-service-account-token',
 });
 
@@ -139,8 +126,8 @@ import { pairwise, map } from 'rxjs';
 
 // Initialize the SDK
 const sdk = new PromptQLSdk({
-  projectId: 'your-project-id',
-  serviceAccountToken: 'your-service-account-token',
+  baseUrl: 'https://your-data-plane-url',
+  accessToken: 'personal-access-token',
 });
 
 // Start a new thread
@@ -260,49 +247,38 @@ console.log(source.replaceAll(xmlRegex, ""));
 ## Configuration Options
 
 ```typescript
-interface PromptQLSdkOptions {
-  // Required: Project ID from PromptQL console
-  projectId: string;
-
-  // Required: Service account token for authentication
-  serviceAccountToken: string;
-
-  // Optional: Base URL of PromptQL data plane
-  // Default: 'https://promptql.ddn.hasura.app'
-  promptqlBaseUrl?: string;
-
-  // Optional: Custom authentication host (for self-hosted control plane)
-  authHost?: string;
-
-  // Optional: Custom fetch implementation
+export type PromptQLSdkOptions = {
+  /**
+   * Base URL of the PromptQL API.
+   */
+  baseUrl?: string;
+  /**
+   * Access token of the PromptQL project.
+   */
+  accessToken: string;
+  /**
+   * A function to use instead of calling the Fetch API directly when sending HTTP requests to your GraphQL endpoint. The function must conform to the signature of fetch.
+   * By default, the Fetch API is used unless it isn't available in your runtime environment.
+   */
   fetch?: typeof fetch;
-
-  // Optional: Additional headers for all requests
+  /**
+   * An object representing headers to include in every HTTP request.
+   */
   headers?: Record<string, string>;
-
-  // Optional: IANA timezone for time-based queries
-  // Default: Client's timezone
+  /**
+   * An [IANA timezone](https://data.iana.org/time-zones/tzdb-2021a/zone1970.tab) for interpreting time-based queries. Default is the timezone from the client config.
+   */
   timezone?: string;
-
-  // Optional: DDN build ID
-  // Default: Applied build
-  buildId?: string;
-}
+};
 ```
 
-## License
-
-Apache License 2.0 - see [LICENSE](LICENSE) file for details.
-
 ## Links
 
-- [PromptQL Website](https://promptql.hasura.io)
-- [PromptQL Console](https://promptql.console.hasura.io)
-- [Hasura Documentation](https://hasura.io/docs)
+- [PromptQL Website](https://promptql.io)
+- [PromptQL App](https://hasura.ql.app)
 
 ## Support
 
 For support, please:
 
-1. Check the [documentation](https://promptql.hasura.io)
-2. Join the [Hasura Discord community](https://discord.gg/hasura)
+1. Join the [Hasura Discord community](https://discord.gg/hasura)

package/biome.json

@@ -1,5 +1,5 @@
 {
-	"$schema": "https://biomejs.dev/schemas/2.4.13/schema.json",
+	"$schema": "https://biomejs.dev/schemas/2.5.1/schema.json",
 	"root": false,
 	"files": {
 		"includes": [