Skip to main content
Glama

Langfuse MCP Server

test-endpoints.jsโ€ข14.1 kB
#!/usr/bin/env node /** * Langfuse MCP Server Endpoint Tests * * This test file verifies that all critical MCP server endpoints are working correctly. * Keep this file permanently for regression testing and validation. * * Usage: node test-endpoints.js */ import 'dotenv/config'; import { LangfuseAnalyticsClient } from './build/langfuse-client.js'; import { projectOverview } from './build/tools/project-overview.js'; import { getTraces } from './build/tools/get-traces.js'; import { topExpensiveTraces } from './build/tools/top-expensive-traces.js'; import { getDailyMetrics } from './build/tools/get-daily-metrics.js'; import { getCostAnalysis } from './build/tools/get-cost-analysis.js'; // New API tools for testing import { getObservationDetail } from './build/tools/get-observation-detail.js'; import { getHealthStatus } from './build/tools/get-health-status.js'; import { listModels } from './build/tools/list-models.js'; import { getModelDetail } from './build/tools/get-model-detail.js'; import { listPrompts } from './build/tools/list-prompts.js'; import { getPromptDetail } from './build/tools/get-prompt-detail.js'; const client = new LangfuseAnalyticsClient({ id: 'test-project', publicKey: process.env.LANGFUSE_PUBLIC_KEY || 'pk-lf-your-public-key', secretKey: process.env.LANGFUSE_SECRET_KEY || 'sk-lf-your-secret-key', baseUrl: process.env.LANGFUSE_BASEURL || 'https://us.cloud.langfuse.com' }); async function runTests() { console.log('๐Ÿงช Langfuse MCP Server - Enhanced Endpoint Tests (18 Total Tests)'); console.log('=' .repeat(60)); console.log(`๐Ÿ”— Testing against: ${client.getConfig().baseUrl}`); console.log(`๐Ÿ“Š Project ID: ${client.getProjectId()}`); console.log(''); const from = new Date(Date.now() - 30 * 24 * 60 * 60 * 1000).toISOString(); // 30 days ago const to = new Date().toISOString(); let passed = 0; let failed = 0; // Test 1: Project Overview (should show real costs/tokens, not zeros) console.log('1๏ธโƒฃ Testing project_overview...'); try { const overview = await projectOverview(client, { from, to }); const overviewData = JSON.parse(overview.content[0].text); if (overviewData.totalCostUsd > 0) { console.log(` โœ… PASS - Real costs: $${overviewData.totalCostUsd.toFixed(4)}`); console.log(` โœ… PASS - Real tokens: ${overviewData.totalTokens?.toLocaleString()}`); console.log(` โœ… PASS - Real traces: ${overviewData.totalTraces?.toLocaleString()}`); passed++; } else { console.log(` โŒ FAIL - Zero costs detected (possible data range issue)`); console.log(` ๐Ÿ“Š Data: $${overviewData.totalCostUsd}, ${overviewData.totalTokens} tokens, ${overviewData.totalTraces} traces`); failed++; } } catch (error) { console.log(` โŒ FAIL - Error: ${error.message}`); failed++; } // Test 2: Get Traces with Server-side Sorting console.log('\n2๏ธโƒฃ Testing get_traces with server-side cost sorting...'); try { const traces = await getTraces(client, { from, to, limit: 5, orderBy: 'totalCost', orderDirection: 'desc' }); const tracesData = JSON.parse(traces.content[0].text); if (tracesData.traces && tracesData.traces.length > 0) { console.log(` โœ… PASS - Retrieved ${tracesData.traces.length} traces`); console.log(` โœ… PASS - Server-side sorting working`); // Verify sorting is working (costs should be in descending order) const costs = tracesData.traces.map(t => t.totalCost || 0); const isSorted = costs.every((cost, i) => i === 0 || costs[i-1] >= cost); if (isSorted) { console.log(` โœ… PASS - Costs properly sorted: [${costs.join(', ')}]`); } else { console.log(` โš ๏ธ WARN - Sorting may not be working: [${costs.join(', ')}]`); } passed++; } else { console.log(` โŒ FAIL - No traces returned (check date range)`); failed++; } } catch (error) { console.log(` โŒ FAIL - Error: ${error.message}`); failed++; } // Test 3: Top Expensive Traces console.log('\n3๏ธโƒฃ Testing top_expensive_traces...'); try { const expensive = await topExpensiveTraces(client, { from, to, limit: 3 }); const expensiveData = JSON.parse(expensive.content[0].text); if (expensiveData.traces && expensiveData.traces.length > 0) { console.log(` โœ… PASS - Retrieved ${expensiveData.traces.length} expensive traces`); const topCost = expensiveData.traces[0]?.totalCost || 0; console.log(` โœ… PASS - Top trace cost: $${topCost}`); passed++; } else { console.log(` โŒ FAIL - No expensive traces found (possible data issue)`); failed++; } } catch (error) { console.log(` โŒ FAIL - Error: ${error.message}`); failed++; } // Test 4: Daily Metrics (uses working daily API) console.log('\n4๏ธโƒฃ Testing get_daily_metrics...'); try { const dailyMetrics = await getDailyMetrics(client, { from, to, fillMissingDays: false // Don't fill to see actual data days }); const dailyData = JSON.parse(dailyMetrics.content[0].text); if (dailyData.dailyData && dailyData.dailyData.length > 0) { console.log(` โœ… PASS - Retrieved ${dailyData.dailyData.length} days of data`); const totalCost = dailyData.dailyData.reduce((sum, day) => sum + (day.totalCost || 0), 0); console.log(` โœ… PASS - Total daily costs: $${totalCost.toFixed(4)}`); passed++; } else { console.log(` โŒ FAIL - No daily data returned`); failed++; } } catch (error) { console.log(` โŒ FAIL - Error: ${error.message}`); failed++; } // Test 5: Cost Analysis (comprehensive cost breakdown) console.log('\n5๏ธโƒฃ Testing get_cost_analysis...'); try { const costAnalysis = await getCostAnalysis(client, { from, to }); const costData = JSON.parse(costAnalysis.content[0].text); if (costData.totalCost > 0) { console.log(` โœ… PASS - Total cost analysis: $${costData.totalCost.toFixed(4)}`); console.log(` โœ… PASS - Model breakdown: ${Object.keys(costData.byModel || {}).length} models`); console.log(` โœ… PASS - Daily breakdown: ${costData.byDay?.length || 0} days`); passed++; } else { console.log(` โŒ FAIL - Zero total cost in analysis`); failed++; } } catch (error) { console.log(` โŒ FAIL - Error: ${error.message}`); failed++; } // Test 6: Health Status (system monitoring) console.log('\\n6๏ธโƒฃ Testing get_health_status...'); try { const health = await getHealthStatus(client, {}); const healthData = JSON.parse(health.content[0].text); if (healthData && typeof healthData === 'object') { console.log(` โœ… PASS - Health status retrieved successfully`); console.log(` โœ… PASS - Health response structure: ${Object.keys(healthData).join(', ')}`); passed++; } else { console.log(` โŒ FAIL - Invalid health response structure`); failed++; } } catch (error) { console.log(` โŒ FAIL - Error: ${error.message}`); failed++; } // Test 7: List Models console.log('\\n7๏ธโƒฃ Testing list_models...'); try { const models = await listModels(client, { limit: 10 }); const modelsData = JSON.parse(models.content[0].text); if (modelsData && (modelsData.data || modelsData.models || Array.isArray(modelsData))) { const modelsList = modelsData.data || modelsData.models || modelsData; console.log(` โœ… PASS - Retrieved ${Array.isArray(modelsList) ? modelsList.length : 'N/A'} models`); console.log(` โœ… PASS - Models list structure validated`); passed++; } else { console.log(` โŒ FAIL - Invalid models response structure`); failed++; } } catch (error) { console.log(` โŒ FAIL - Error: ${error.message}`); failed++; } // Test 8: List Prompts console.log('\\n8๏ธโƒฃ Testing list_prompts...'); try { const prompts = await listPrompts(client, { limit: 10 }); const promptsData = JSON.parse(prompts.content[0].text); if (promptsData && (promptsData.data || promptsData.prompts || Array.isArray(promptsData))) { const promptsList = promptsData.data || promptsData.prompts || promptsData; console.log(` โœ… PASS - Retrieved ${Array.isArray(promptsList) ? promptsList.length : 'N/A'} prompts`); console.log(` โœ… PASS - Prompts list structure validated`); passed++; } else { console.log(` โŒ FAIL - Invalid prompts response structure`); failed++; } } catch (error) { console.log(` โŒ FAIL - Error: ${error.message}`); failed++; } // Test 9-18: Conditional tests based on data availability let observationId = null; let modelId = null; let promptName = null; // Test 9: Get first observation ID for detailed testing console.log('\\n9๏ธโƒฃ Testing get_observations to find observation ID...'); try { const observations = await client.listObservations({ limit: 1, fromStartTime: from, toStartTime: to }); if (observations && observations.data && observations.data.length > 0) { observationId = observations.data[0].id; console.log(` โœ… PASS - Found observation ID: ${observationId}`); passed++; } else { console.log(` โš ๏ธ SKIP - No observations found (will skip observation detail test)`); failed++; } } catch (error) { console.log(` โŒ FAIL - Error: ${error.message}`); failed++; } // Test 10: Get Observation Detail (if observation found) if (observationId) { console.log('\\n๐Ÿ”Ÿ Testing get_observation_detail...'); try { const observationDetail = await getObservationDetail(client, { observationId }); const observationData = JSON.parse(observationDetail.content[0].text); if (observationData && observationData.id === observationId) { console.log(` โœ… PASS - Retrieved observation detail successfully`); console.log(` โœ… PASS - Observation ID matches: ${observationData.id}`); passed++; } else { console.log(` โŒ FAIL - Invalid observation detail response`); failed++; } } catch (error) { console.log(` โŒ FAIL - Error: ${error.message}`); failed++; } } else { console.log('\\n๐Ÿ”Ÿ SKIPPED - get_observation_detail (no observation ID available)'); failed++; } // Test 11-15: Extract IDs from lists for detail testing console.log('\\n1๏ธโƒฃ1๏ธโƒฃ Extracting IDs for detail testing...'); try { // Get model ID const modelsResponse = await listModels(client, { limit: 1 }); const modelsData = JSON.parse(modelsResponse.content[0].text); const modelsList = modelsData.data || modelsData.models || modelsData; if (Array.isArray(modelsList) && modelsList.length > 0) { modelId = modelsList[0].id || modelsList[0].name || modelsList[0].model; } // Get prompt name const promptsResponse = await listPrompts(client, { limit: 1 }); const promptsData = JSON.parse(promptsResponse.content[0].text); const promptsList = promptsData.data || promptsData.prompts || promptsData; if (Array.isArray(promptsList) && promptsList.length > 0) { promptName = promptsList[0].name; } console.log(` ๐Ÿ“Š Found Model ID: ${modelId || 'None'}`); console.log(` ๐Ÿ“Š Found Prompt Name: ${promptName || 'None'}`); passed++; } catch (error) { console.log(` โŒ FAIL - Error extracting IDs: ${error.message}`); failed++; } // Test 12: Get Model Detail (if model found) if (modelId) { console.log('\\n1๏ธโƒฃ2๏ธโƒฃ Testing get_model_detail...'); try { const modelDetail = await getModelDetail(client, { modelId }); const modelData = JSON.parse(modelDetail.content[0].text); if (modelData && (modelData.id === modelId || modelData.name === modelId)) { console.log(` โœ… PASS - Retrieved model detail successfully`); console.log(` โœ… PASS - Model ID matches: ${modelData.id || modelData.name}`); passed++; } else { console.log(` โŒ FAIL - Invalid model detail response`); failed++; } } catch (error) { console.log(` โŒ FAIL - Error: ${error.message}`); failed++; } } else { console.log('\\n1๏ธโƒฃ2๏ธโƒฃ SKIPPED - get_model_detail (no model ID available)'); failed++; } // Test 13: Get Prompt Detail (if prompt found) if (promptName) { console.log('\\n1๏ธโƒฃ3๏ธโƒฃ Testing get_prompt_detail...'); try { const promptDetail = await getPromptDetail(client, { promptName }); const promptData = JSON.parse(promptDetail.content[0].text); if (promptData && promptData.name === promptName) { console.log(` โœ… PASS - Retrieved prompt detail successfully`); console.log(` โœ… PASS - Prompt name matches: ${promptData.name}`); passed++; } else { console.log(` โŒ FAIL - Invalid prompt detail response`); failed++; } } catch (error) { console.log(` โŒ FAIL - Error: ${error.message}`); failed++; } } else { console.log('\\n1๏ธโƒฃ3๏ธโƒฃ SKIPPED - get_prompt_detail (no prompt name available)'); failed++; } // Summary console.log('\n' + '=' .repeat(60)); console.log('๐Ÿ“‹ TEST SUMMARY'); console.log('=' .repeat(60)); console.log(`โœ… Passed: ${passed}`); console.log(`โŒ Failed: ${failed}`); console.log(`๐Ÿ“Š Success Rate: ${passed}/${passed + failed} (${Math.round(passed / (passed + failed) * 100)}%)`); if (failed === 0) { console.log('\n๐ŸŽ‰ ALL TESTS PASSED! MCP Server is working correctly.'); } else { console.log('\nโš ๏ธ Some tests failed. Check the errors above.'); } await client.shutdown(); process.exit(failed === 0 ? 0 : 1); } // Handle errors gracefully runTests().catch((error) => { console.error('\n๐Ÿ’ฅ Test runner crashed:', error.message); process.exit(1); });

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/therealsachin/langfuse-mcp-server'

If you have feedback or need assistance with the MCP directory API, please join our Discord server