@kaiban/sdk 0.1.1

package/dist/lib/resources/BenchmarkExecutionsClient.d.ts

@@ -0,0 +1,250 @@
+import { HttpClient } from '../http/HttpClient';
+import { RequestOptions } from '../http/types';
+import { BenchmarkExecution, CreateBenchmarkExecutionInput, TestCaseResult, CreateTestCaseResultInput, UpdateBenchmarkExecutionInput, UpdateTestCaseResultInput } from '../../types/entities';
+import { ListParams, Paginated } from '../../types/responses';
+/**
+ * Client for managing benchmark executions and test case results
+ * Benchmark executions track the results of running benchmarks against agents
+ * @category Resources
+ */
+export declare class BenchmarkExecutionsClient {
+    private readonly http;
+    constructor(http: HttpClient);
+    /**
+     * List benchmark executions with pagination, filters, and sorting
+     *
+     * @param params - Optional query parameters for filtering and pagination
+     * @param params.cursor - Cursor for the next page of results
+     * @param params.limit - Number of items per page
+     * @param params.order_by - Fields to sort by
+     * @param options - Optional request configuration
+     *
+     * @returns A paginated list of benchmark executions
+     *
+     * @example
+     * ```typescript
+     * const result = await client.benchmarkExecutions.list({ limit: 10 });
+     * console.log(result.items);
+     * ```
+     */
+    list(params?: ListParams, options?: RequestOptions): Promise<Paginated<BenchmarkExecution>>;
+    /**
+     * Iterate through all benchmark executions using async generator
+     * Automatically handles pagination by following next_page_token
+     *
+     * @param params - Optional query parameters (excluding cursor)
+     * @param options - Optional request configuration
+     *
+     * @yields Individual BenchmarkExecution objects
+     *
+     * @example
+     * ```typescript
+     * for await (const execution of client.benchmarkExecutions.listAll()) {
+     *   console.log(execution.id, execution.status);
+     * }
+     * ```
+     */
+    listAll(params?: Omit<ListParams, 'cursor'>, options?: RequestOptions): AsyncGenerator<BenchmarkExecution, void, unknown>;
+    /**
+     * Create a new benchmark execution
+     *
+     * @param data - Execution creation data
+     * @param options - Optional request configuration
+     *
+     * @returns The newly created benchmark execution
+     *
+     * @throws {Error} If creation fails
+     *
+     * @example
+     * ```typescript
+     * const execution = await client.benchmarkExecutions.create({
+     *   benchmark_id: 'benchmark-123',
+     *   agent_id: 'agent-456',
+     *   status: 'pending'
+     * });
+     * ```
+     */
+    create(data: CreateBenchmarkExecutionInput, options?: RequestOptions): Promise<BenchmarkExecution>;
+    /**
+     * Get a single benchmark execution by ID
+     *
+     * @param id - The unique identifier of the benchmark execution
+     * @param options - Optional request configuration
+     *
+     * @returns The benchmark execution object
+     *
+     * @throws {Error} If the execution is not found or request fails
+     *
+     * @example
+     * ```typescript
+     * const execution = await client.benchmarkExecutions.get('execution-123');
+     * console.log(execution.status, execution.results);
+     * ```
+     */
+    get(id: string, options?: RequestOptions): Promise<BenchmarkExecution>;
+    /**
+     * Update a benchmark execution's properties
+     *
+     * @param id - The unique identifier of the execution to update
+     * @param data - Execution data with fields to update
+     * @param options - Optional request configuration
+     *
+     * @returns The updated benchmark execution object
+     *
+     * @throws {Error} If the execution is not found or update fails
+     *
+     * @example
+     * ```typescript
+     * const updated = await client.benchmarkExecutions.update('execution-123', {
+     *   status: 'completed'
+     * });
+     * ```
+     */
+    update(id: string, data: UpdateBenchmarkExecutionInput, options?: RequestOptions): Promise<BenchmarkExecution>;
+    /**
+     * Delete a benchmark execution by ID
+     *
+     * @param id - The unique identifier of the execution to delete
+     * @param options - Optional request configuration
+     *
+     * @returns Promise that resolves when the execution is deleted
+     *
+     * @throws {Error} If the execution is not found or deletion fails
+     *
+     * @example
+     * ```typescript
+     * await client.benchmarkExecutions.delete('execution-123');
+     * ```
+     */
+    delete(id: string, options?: RequestOptions): Promise<void>;
+    /**
+     * Create a test case result for a benchmark execution
+     *
+     * @param params - Test case creation parameters
+     * @param params.execution_id - The unique identifier of the benchmark execution
+     * @param params.body - Test case result data
+     * @param options - Optional request configuration
+     *
+     * @returns The newly created test case result
+     *
+     * @throws {Error} If the execution is not found or creation fails
+     *
+     * @example
+     * ```typescript
+     * const testCase = await client.benchmarkExecutions.createTestCase({
+     *   execution_id: 'execution-123',
+     *   body: {
+     *     test_case_id: 'test-1',
+     *     status: 'passed',
+     *     score: 95
+     *   }
+     * });
+     * ```
+     */
+    createTestCase(params: {
+        execution_id: string;
+        body: CreateTestCaseResultInput;
+    }, options?: RequestOptions): Promise<TestCaseResult>;
+    /**
+     * List test case results for a benchmark execution
+     *
+     * @param params - List parameters including execution_id and pagination options
+     * @param params.execution_id - The unique identifier of the benchmark execution
+     * @param params.cursor - Optional cursor for the next page of results
+     * @param params.limit - Optional number of items per page
+     * @param options - Optional request configuration
+     *
+     * @returns A paginated list of test case results
+     *
+     * @example
+     * ```typescript
+     * const results = await client.benchmarkExecutions.listTestCases({
+     *   execution_id: 'execution-123',
+     *   limit: 20
+     * });
+     * console.log(results.items);
+     * ```
+     */
+    listTestCases(params: {
+        execution_id: string;
+    } & ListParams, options?: RequestOptions): Promise<Paginated<TestCaseResult>>;
+    /**
+     * Get a single test case result by ID
+     *
+     * @param params - Parameters for getting the test case
+     * @param params.execution_id - The unique identifier of the benchmark execution
+     * @param params.test_case_id - The unique identifier of the test case result
+     * @param options - Optional request configuration
+     *
+     * @returns The test case result object
+     *
+     * @throws {Error} If the test case is not found or request fails
+     *
+     * @example
+     * ```typescript
+     * const testCase = await client.benchmarkExecutions.getTestCase({
+     *   execution_id: 'execution-123',
+     *   test_case_id: 'test-1'
+     * });
+     * console.log(testCase.status, testCase.score);
+     * ```
+     */
+    getTestCase(params: {
+        execution_id: string;
+        test_case_id: string;
+    }, options?: RequestOptions): Promise<TestCaseResult>;
+    /**
+     * Update a test case result's properties
+     *
+     * @param params - Update parameters
+     * @param params.execution_id - The unique identifier of the benchmark execution
+     * @param params.test_case_id - The unique identifier of the test case result
+     * @param params.body - Test case result data with fields to update
+     * @param options - Optional request configuration
+     *
+     * @returns The updated test case result object
+     *
+     * @throws {Error} If the test case is not found or update fails
+     *
+     * @example
+     * ```typescript
+     * const updated = await client.benchmarkExecutions.updateTestCase({
+     *   execution_id: 'execution-123',
+     *   test_case_id: 'test-1',
+     *   body: {
+     *     status: 'passed',
+     *     score: 98
+     *   }
+     * });
+     * ```
+     */
+    updateTestCase(params: {
+        execution_id: string;
+        test_case_id: string;
+        body: UpdateTestCaseResultInput;
+    }, options?: RequestOptions): Promise<TestCaseResult>;
+    /**
+     * Delete a test case result by ID
+     *
+     * @param params - Delete parameters
+     * @param params.execution_id - The unique identifier of the benchmark execution
+     * @param params.test_case_id - The unique identifier of the test case result
+     * @param options - Optional request configuration
+     *
+     * @returns Promise that resolves when the test case is deleted
+     *
+     * @throws {Error} If the test case is not found or deletion fails
+     *
+     * @example
+     * ```typescript
+     * await client.benchmarkExecutions.deleteTestCase({
+     *   execution_id: 'execution-123',
+     *   test_case_id: 'test-1'
+     * });
+     * ```
+     */
+    deleteTestCase(params: {
+        execution_id: string;
+        test_case_id: string;
+    }, options?: RequestOptions): Promise<void>;
+}

package/dist/lib/resources/BenchmarkExecutionsClient.js

@@ -0,0 +1,261 @@
+/**
+ * Client for managing benchmark executions and test case results
+ * Benchmark executions track the results of running benchmarks against agents
+ * @category Resources
+ */
+export class BenchmarkExecutionsClient {
+    constructor(http) {
+        this.http = http;
+    }
+    /**
+     * List benchmark executions with pagination, filters, and sorting
+     *
+     * @param params - Optional query parameters for filtering and pagination
+     * @param params.cursor - Cursor for the next page of results
+     * @param params.limit - Number of items per page
+     * @param params.order_by - Fields to sort by
+     * @param options - Optional request configuration
+     *
+     * @returns A paginated list of benchmark executions
+     *
+     * @example
+     * ```typescript
+     * const result = await client.benchmarkExecutions.list({ limit: 10 });
+     * console.log(result.items);
+     * ```
+     */
+    list(params, options) {
+        return this.http.list('/benchmark-executions', params, options);
+    }
+    /**
+     * Iterate through all benchmark executions using async generator
+     * Automatically handles pagination by following next_page_token
+     *
+     * @param params - Optional query parameters (excluding cursor)
+     * @param options - Optional request configuration
+     *
+     * @yields Individual BenchmarkExecution objects
+     *
+     * @example
+     * ```typescript
+     * for await (const execution of client.benchmarkExecutions.listAll()) {
+     *   console.log(execution.id, execution.status);
+     * }
+     * ```
+     */
+    async *listAll(params, options) {
+        let pageToken = undefined;
+        do {
+            const page = await this.list({ ...(params || {}), cursor: pageToken }, options);
+            for (const item of page.items)
+                yield item;
+            pageToken = page.next_page_token;
+        } while (pageToken);
+    }
+    /**
+     * Create a new benchmark execution
+     *
+     * @param data - Execution creation data
+     * @param options - Optional request configuration
+     *
+     * @returns The newly created benchmark execution
+     *
+     * @throws {Error} If creation fails
+     *
+     * @example
+     * ```typescript
+     * const execution = await client.benchmarkExecutions.create({
+     *   benchmark_id: 'benchmark-123',
+     *   agent_id: 'agent-456',
+     *   status: 'pending'
+     * });
+     * ```
+     */
+    create(data, options) {
+        return this.http.post('/benchmark-executions', data, options);
+    }
+    /**
+     * Get a single benchmark execution by ID
+     *
+     * @param id - The unique identifier of the benchmark execution
+     * @param options - Optional request configuration
+     *
+     * @returns The benchmark execution object
+     *
+     * @throws {Error} If the execution is not found or request fails
+     *
+     * @example
+     * ```typescript
+     * const execution = await client.benchmarkExecutions.get('execution-123');
+     * console.log(execution.status, execution.results);
+     * ```
+     */
+    get(id, options) {
+        return this.http.get(`/benchmark-execution/${encodeURIComponent(id)}`, options);
+    }
+    /**
+     * Update a benchmark execution's properties
+     *
+     * @param id - The unique identifier of the execution to update
+     * @param data - Execution data with fields to update
+     * @param options - Optional request configuration
+     *
+     * @returns The updated benchmark execution object
+     *
+     * @throws {Error} If the execution is not found or update fails
+     *
+     * @example
+     * ```typescript
+     * const updated = await client.benchmarkExecutions.update('execution-123', {
+     *   status: 'completed'
+     * });
+     * ```
+     */
+    update(id, data, options) {
+        return this.http.put(`/benchmark-execution/${encodeURIComponent(id)}`, data, options);
+    }
+    /**
+     * Delete a benchmark execution by ID
+     *
+     * @param id - The unique identifier of the execution to delete
+     * @param options - Optional request configuration
+     *
+     * @returns Promise that resolves when the execution is deleted
+     *
+     * @throws {Error} If the execution is not found or deletion fails
+     *
+     * @example
+     * ```typescript
+     * await client.benchmarkExecutions.delete('execution-123');
+     * ```
+     */
+    delete(id, options) {
+        return this.http.delete(`/benchmark-execution/${encodeURIComponent(id)}`, options);
+    }
+    /**
+     * Create a test case result for a benchmark execution
+     *
+     * @param params - Test case creation parameters
+     * @param params.execution_id - The unique identifier of the benchmark execution
+     * @param params.body - Test case result data
+     * @param options - Optional request configuration
+     *
+     * @returns The newly created test case result
+     *
+     * @throws {Error} If the execution is not found or creation fails
+     *
+     * @example
+     * ```typescript
+     * const testCase = await client.benchmarkExecutions.createTestCase({
+     *   execution_id: 'execution-123',
+     *   body: {
+     *     test_case_id: 'test-1',
+     *     status: 'passed',
+     *     score: 95
+     *   }
+     * });
+     * ```
+     */
+    createTestCase(params, options) {
+        return this.http.post(`/benchmark-execution/${encodeURIComponent(params.execution_id)}/test-cases`, params.body, options);
+    }
+    /**
+     * List test case results for a benchmark execution
+     *
+     * @param params - List parameters including execution_id and pagination options
+     * @param params.execution_id - The unique identifier of the benchmark execution
+     * @param params.cursor - Optional cursor for the next page of results
+     * @param params.limit - Optional number of items per page
+     * @param options - Optional request configuration
+     *
+     * @returns A paginated list of test case results
+     *
+     * @example
+     * ```typescript
+     * const results = await client.benchmarkExecutions.listTestCases({
+     *   execution_id: 'execution-123',
+     *   limit: 20
+     * });
+     * console.log(results.items);
+     * ```
+     */
+    listTestCases(params, options) {
+        const { execution_id, ...rest } = params;
+        return this.http.get(`/benchmark-execution/${encodeURIComponent(execution_id)}/test-cases`, { ...options, query: { ...(rest || {}) } });
+    }
+    /**
+     * Get a single test case result by ID
+     *
+     * @param params - Parameters for getting the test case
+     * @param params.execution_id - The unique identifier of the benchmark execution
+     * @param params.test_case_id - The unique identifier of the test case result
+     * @param options - Optional request configuration
+     *
+     * @returns The test case result object
+     *
+     * @throws {Error} If the test case is not found or request fails
+     *
+     * @example
+     * ```typescript
+     * const testCase = await client.benchmarkExecutions.getTestCase({
+     *   execution_id: 'execution-123',
+     *   test_case_id: 'test-1'
+     * });
+     * console.log(testCase.status, testCase.score);
+     * ```
+     */
+    getTestCase(params, options) {
+        return this.http.get(`/benchmark-execution/${encodeURIComponent(params.execution_id)}/test-cases/${encodeURIComponent(params.test_case_id)}`, options);
+    }
+    /**
+     * Update a test case result's properties
+     *
+     * @param params - Update parameters
+     * @param params.execution_id - The unique identifier of the benchmark execution
+     * @param params.test_case_id - The unique identifier of the test case result
+     * @param params.body - Test case result data with fields to update
+     * @param options - Optional request configuration
+     *
+     * @returns The updated test case result object
+     *
+     * @throws {Error} If the test case is not found or update fails
+     *
+     * @example
+     * ```typescript
+     * const updated = await client.benchmarkExecutions.updateTestCase({
+     *   execution_id: 'execution-123',
+     *   test_case_id: 'test-1',
+     *   body: {
+     *     status: 'passed',
+     *     score: 98
+     *   }
+     * });
+     * ```
+     */
+    updateTestCase(params, options) {
+        return this.http.put(`/benchmark-execution/${encodeURIComponent(params.execution_id)}/test-cases/${encodeURIComponent(params.test_case_id)}`, params.body, options);
+    }
+    /**
+     * Delete a test case result by ID
+     *
+     * @param params - Delete parameters
+     * @param params.execution_id - The unique identifier of the benchmark execution
+     * @param params.test_case_id - The unique identifier of the test case result
+     * @param options - Optional request configuration
+     *
+     * @returns Promise that resolves when the test case is deleted
+     *
+     * @throws {Error} If the test case is not found or deletion fails
+     *
+     * @example
+     * ```typescript
+     * await client.benchmarkExecutions.deleteTestCase({
+     *   execution_id: 'execution-123',
+     *   test_case_id: 'test-1'
+     * });
+     * ```
+     */
+    deleteTestCase(params, options) {
+        return this.http.delete(`/benchmark-execution/${encodeURIComponent(params.execution_id)}/test-cases/${encodeURIComponent(params.test_case_id)}`, options);
+    }
+}

package/dist/lib/resources/BenchmarksClient.d.ts

@@ -0,0 +1,159 @@
+import { HttpClient } from '../http/HttpClient';
+import { RequestOptions } from '../http/types';
+import { Benchmark, CreateBenchmarkInput, UpdateBenchmarkInput } from '../../types/entities';
+import { ListParams, Paginated } from '../../types/responses';
+/**
+ * Client for managing benchmarks in the Kaiban platform
+ * Benchmarks are used to evaluate and test agent performance
+ * @category Resources
+ */
+export declare class BenchmarksClient {
+    private readonly http;
+    constructor(http: HttpClient);
+    /**
+     * List benchmarks with pagination, filters, and sorting
+     *
+     * @param params - Optional query parameters for filtering and pagination
+     * @param params.cursor - Cursor for the next page of results
+     * @param params.limit - Number of items per page
+     * @param params.order_by - Fields to sort by
+     * @param options - Optional request configuration
+     *
+     * @returns A paginated list of benchmarks
+     *
+     * @example
+     * ```typescript
+     * const result = await client.benchmarks.list({ limit: 10 });
+     * console.log(result.items);
+     * ```
+     */
+    list(params?: ListParams, options?: RequestOptions): Promise<Paginated<Benchmark>>;
+    /**
+     * Iterate through all benchmarks using async generator
+     * Automatically handles pagination by following next_page_token
+     *
+     * @param params - Optional query parameters (excluding cursor)
+     * @param options - Optional request configuration
+     *
+     * @yields Individual Benchmark objects
+     *
+     * @example
+     * ```typescript
+     * for await (const benchmark of client.benchmarks.listAll()) {
+     *   console.log(benchmark.id, benchmark.name);
+     * }
+     * ```
+     */
+    listAll(params?: Omit<ListParams, 'cursor'>, options?: RequestOptions): AsyncGenerator<Benchmark, void, unknown>;
+    /**
+     * Get a single benchmark by ID
+     *
+     * @param id - The unique identifier of the benchmark
+     * @param options - Optional request configuration
+     *
+     * @returns The benchmark object
+     *
+     * @throws {Error} If the benchmark is not found or request fails
+     *
+     * @example
+     * ```typescript
+     * const benchmark = await client.benchmarks.get('benchmark-123');
+     * console.log(benchmark.name, benchmark.description);
+     * ```
+     */
+    get(id: string, options?: RequestOptions): Promise<Benchmark>;
+    /**
+     * Create a new benchmark
+     *
+     * @param data - Benchmark creation data
+     * @param data.name - The benchmark name
+     * @param data.description - Optional description
+     * @param options - Optional request configuration
+     *
+     * @returns The newly created benchmark
+     *
+     * @throws {Error} If creation fails
+     *
+     * @example
+     * ```typescript
+     * const benchmark = await client.benchmarks.create({
+     *   name: 'Performance Test Suite',
+     *   description: 'Test agent response times',
+     *   test_cases: [...]
+     * });
+     * ```
+     */
+    create(data: CreateBenchmarkInput, options?: RequestOptions): Promise<Benchmark>;
+    /**
+     * Update a benchmark's properties
+     *
+     * @param id - The unique identifier of the benchmark to update
+     * @param data - Benchmark data with fields to update
+     * @param options - Optional request configuration
+     *
+     * @returns The updated benchmark object
+     *
+     * @throws {Error} If the benchmark is not found or update fails
+     *
+     * @example
+     * ```typescript
+     * const updated = await client.benchmarks.update('benchmark-123', {
+     *   name: 'Updated Benchmark Name'
+     * });
+     * ```
+     */
+    update(id: string, data: UpdateBenchmarkInput, options?: RequestOptions): Promise<Benchmark>;
+    /**
+     * Execute a benchmark against an agent
+     * Runs the benchmark test suite with specified configuration
+     *
+     * @param params - Execution parameters
+     * @param params.benchmark_id - The unique identifier of the benchmark to execute
+     * @param params.body - Execution configuration
+     * @param params.body.agentConfig - Agent configuration including ID, name, and card URL
+     * @param params.body.thresholds - Optional performance thresholds
+     * @param params.body.maxTests - Optional maximum number of tests to run
+     * @param params.body.parallel - Optional flag to run tests in parallel
+     * @param params.body.batchSize - Optional batch size for parallel execution
+     * @param options - Optional request configuration
+     *
+     * @returns A response with execution status message
+     *
+     * @throws {Error} If the benchmark is not found or execution fails
+     *
+     * @example
+     * ```typescript
+     * const result = await client.benchmarks.execute({
+     *   benchmark_id: 'benchmark-123',
+     *   body: {
+     *     agentConfig: {
+     *       id: 'agent-456',
+     *       name: 'Test Agent',
+     *       cardUrl: 'https://example.com/agent-card'
+     *     },
+     *     maxTests: 100,
+     *     parallel: true,
+     *     batchSize: 10
+     *   }
+     * });
+     * console.log(result.message);
+     * ```
+     */
+    execute(params: {
+        benchmark_id: string;
+        body: {
+            agentConfig: {
+                id: string;
+                name?: string;
+                type?: string;
+                cardUrl?: string;
+            };
+            thresholds?: Record<string, number>;
+            maxTests?: number;
+            parallel?: boolean;
+            batchSize?: number;
+        };
+    }, options?: RequestOptions): Promise<{
+        message: string;
+    }>;
+}