@virtualkitchenco/multiverse-sdk 0.0.26 → 0.0.27

package/README.md

@@ -5,7 +5,13 @@ Simulation testing for AI agents. Test your agent against realistic scenarios wi
 ## Install
 
 ```bash
-npm install @virtualkitchenco/multiverse-sdk
+npm install @virtualkitchenco/multiverse-sdk zod
+```
+
+For LangChain agents:
+
+```bash
+npm install @virtualkitchenco/multiverse-sdk zod @langchain/core @langchain/anthropic
 ```
 
 ## Quick Start
@@ -17,7 +23,7 @@ import { multiverse } from '@virtualkitchenco/multiverse-sdk';
 import { z } from 'zod';
 
 multiverse.configure({
-  baseUrl: 'https://your-multiverse-server.com',
+  baseUrl: process.env.MULTIVERSE_URL,
   apiKey: process.env.MULTIVERSE_API_KEY,
 });
 
@@ -32,7 +38,9 @@ const test = multiverse.describe({
   }),
 });
 
+const scenarios = await test.generateScenarios({ count: 5 });
 const results = await test.run({
+  scenarios,
   success: (world) => world.getCollection('intake_summaries').size > 0,
 });
 
@@ -71,7 +79,7 @@ Initialize the SDK. Call once at startup.
 
 ```typescript
 multiverse.configure({
-  baseUrl: 'https://your-multiverse-server.com',
+  baseUrl: process.env.MULTIVERSE_URL,
   apiKey: process.env.MULTIVERSE_API_KEY,
 });
 ```
@@ -91,7 +99,7 @@ const searchFlights = multiverse.tool({
   }),
   output: SearchResultSchema,
   execute: async (input) => realSearchFlights(input),
-  effects: (output) =>
+  effects: (output, world) =>
     output.flights.map((f) => ({
       operation: 'create' as const,
       collection: 'flights',
@@ -164,15 +172,24 @@ const test = multiverse.describe({
 });
 ```
 
+| Option | Type | Description |
+|--------|------|-------------|
+| `name` | `string` | Agent name for grouping in the dashboard |
+| `task` | `string` | What the agent is being tested on |
+| `agent` | `AgentFn` | Agent function to test |
+| `conversational` | `boolean` | Enable simulated user (chatbots, assistants). Mutually exclusive with `triggerSchema` |
+| `triggerSchema` | `ZodSchema` | Constrains the generated event payload (autonomous agents only) |
+| `variables` | `ZodSchema` | Typed scenario variables accessible in `success()` via `scenario.variables` |
+
 `conversational` and `triggerSchema` are mutually exclusive at the TypeScript level.
 
 **Agent function signature:**
 
 ```typescript
-async function runAgent(context: {
-  userMessage: string;  // Event payload (autonomous) or user message (conversational)
-  runId: string;        // Stable across turns — use for memory/thread scoping
-}): Promise<string>
+async function runAgent(ctx: {
+  userMessage: string;  // Generated event payload (autonomous) or latest user message (conversational)
+  runId: string;        // Stable across turns, use for memory/thread scoping
+}): Promise<unknown>
 ```
 
 ### `test.generateScenarios(options)`
@@ -181,18 +198,10 @@ Generate test scenarios upfront for inspection or reuse.
 
 ```typescript
 const scenarios = await test.generateScenarios({ count: 10 });
-
-// With typed variables for programmatic assertions
-const scenarios = await test.generateScenarios({
-  count: 5,
-  variables: z.object({
-    expectedBookings: z.number().describe(
-      'Total bookings that should be created (e.g. 2 passengers round-trip = 4)'
-    ),
-  }),
-});
 ```
 
+Variables are typed on `multiverse.describe()` via the `variables` option, not here.
+
 ### `test.saveScenarios(scenarios)`
 
 Save generated scenarios for reuse across runs.
@@ -225,14 +234,17 @@ Run tests against the agent.
 
 ```typescript
 const results = await test.run({
+  scenarios,              // From generateScenarios()
   success: (world, trace, scenario) => {
     return world.getCollection('bookings').size === scenario.variables.expectedBookings;
   },
-  scenarios,              // Pre-generated or auto-generated
-  scenarioCount: 5,       // Auto-generate count (ignored if scenarios provided)
   trialsPerScenario: 4,
   maxTurns: 20,           // Max turns per run (conversational agents)
-  qualityThreshold: 70,
+  qualityThreshold: 70,   // Default: 70
+  criteria: [             // Custom quality criteria (default: communication, error_handling, efficiency, accuracy)
+    { name: 'politeness', description: 'Responds politely at all times' },
+  ],
+  skipReport: true,       // Skip LLM report generation
   concurrency: 8,
   onProgress: (p) => console.log(`${p.completed}/${p.total}`),
   ci: {
@@ -281,7 +293,7 @@ const searchFlightsTool = tool(
 // Wrap for simulation
 const searchFlights = wrap(searchFlightsTool, {
   output: SearchResultSchema,
-  effects: (output) =>
+  effects: (output, world) =>
     output.flights.map((f) => ({
       operation: 'create' as const,
       collection: 'flights',

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@virtualkitchenco/multiverse-sdk",
-  "version": "0.0.26",
+  "version": "0.0.27",
   "description": "Simulation testing SDK for AI agents",
   "type": "module",
   "main": "./dist/index.js",
@@ -26,7 +26,7 @@
   "dependencies": {
     "nanoid": "^5.0.4",
     "zod-to-json-schema": "^3.25.1",
-    "@virtualkitchenco/multiverse-types": "0.0.26"
+    "@virtualkitchenco/multiverse-types": "0.0.27"
   },
   "devDependencies": {
     "@types/node": "^20.11.0",