Langfuse Data Handling

Langfuse Data Handling

Overview

Manage the Langfuse data lifecycle: export traces and scores via the API, configure retention policies, handle GDPR data subject requests, anonymize data for analytics, and maintain audit trails.

Prerequisites

• @langfuse/client installed
• Langfuse API keys with appropriate permissions
• Understanding of your compliance requirements (GDPR, SOC2, HIPAA)

Instructions

Step 1: Export Trace Data via API

import { LangfuseClient } from "@langfuse/client";
import { writeFileSync } from "fs";

const langfuse = new LangfuseClient();

async function exportTraces(options: {
  fromDate: string;
  toDate: string;
  outputFile: string;
  includeObservations?: boolean;
}) {
  const allTraces: any[] = [];
  let page = 1;
  let hasMore = true;

  while (hasMore) {
    const result = await langfuse.api.traces.list({
      fromTimestamp: options.fromDate,
      toTimestamp: options.toDate,
      limit: 100,
      page,
    });

    for (const trace of result.data) {
      const exportItem: any = {
        id: trace.id,
        name: trace.name,
        timestamp: trace.timestamp,
        userId: trace.userId,
        sessionId: trace.sessionId,
        metadata: trace.metadata,
        tags: trace.tags,
      };

      if (options.includeObservations) {
        const observations = await langfuse.api.observations.list({
          traceId: trace.id,
        });
        exportItem.observations = observations.data;
      }

      allTraces.push(exportItem);
    }

    hasMore = result.data.length === 100;
    page++;

    // Rate limit respect
    await new Promise((r) => setTimeout(r, 200));
  }

  writeFileSync(options.outputFile, JSON.stringify(allTraces, null, 2));
  console.log(`Exported ${allTraces.length} traces to ${options.outputFile}`);
}

// Usage
await exportTraces({
  fromDate: "2025-01-01T00:00:00Z",
  toDate: "2025-01-31T23:59:59Z",
  outputFile: "traces-january.json",
  includeObservations: true,
});

Step 2: Export Scores

async function exportScores(fromDate: string, outputFile: string) {
  const scores: any[] = [];
  let page = 1;
  let hasMore = true;

  while (hasMore) {
    const result = await langfuse.api.scores.list({
      fromTimestamp: fromDate,
      limit: 100,
      page,
    });

    scores.push(...result.data);
    hasMore = result.data.length === 100;
    page++;
    await new Promise((r) => setTimeout(r, 200));
  }

  writeFileSync(outputFile, JSON.stringify(scores, null, 2));
  console.log(`Exported ${scores.length} scores to ${outputFile}`);
}

Step 3: Data Retention Configuration

Self-hosted: Set retention via environment variable:

# docker-compose.yml
services:
  langfuse:
    environment:
      - LANGFUSE_RETENTION_DAYS=90

Cloud: Programmatic cleanup of old data:

async function enforceRetention(maxAgeDays: number) {
  const cutoff = new Date(Date.now() - maxAgeDays * 86400000).toISOString();

  const oldTraces = await langfuse.api.traces.list({
    toTimestamp: cutoff,
    limit: 100,
  });

  console.log(`Found ${oldTraces.data.length} traces older than ${maxAgeDays} days`);

  for (const trace of oldTraces.data) {
    await langfuse.api.traces.delete(trace.id);
    await new Promise((r) => setTimeout(r, 100)); // Rate limit
  }
}

// Run as cron job
await enforceRetention(90);

Step 4: GDPR Data Subject Requests

// Handle "Right to Access" -- export all data for a user
async function handleAccessRequest(userId: string) {
  const traces = await langfuse.api.traces.list({
    userId,
    limit: 1000,
  });

  const userData = {
    userId,
    exportDate: new Date().toISOString(),
    traceCount: traces.data.length,
    traces: traces.data.map((t) => ({
      id: t.id,
      name: t.name,
      timestamp: t.timestamp,
      input: t.input,
      output: t.output,
      metadata: t.metadata,
    })),
  };

  writeFileSync(`gdpr-export-${userId}.json`, JSON.stringify(userData, null, 2));
  return userData;
}

// Handle "Right to Erasure" -- delete all data for a user
async function handleDeletionRequest(userId: string) {
  const traces = await langfuse.api.traces.list({
    userId,
    limit: 1000,
  });

  let deleted = 0;
  for (const trace of traces.data) {
    await langfuse.api.traces.delete(trace.id);
    deleted++;
    await new Promise((r) => setTimeout(r, 100));
  }

  console.log(`Deleted ${deleted} traces for user ${userId}`);
  return { userId, tracesDeleted: deleted };
}

Step 5: Data Anonymization for Analytics

import crypto from "crypto";

function anonymizeTrace(trace: any): any {
  return {
    ...trace,
    userId: trace.userId ? crypto.createHash("sha256").update(trace.userId).digest("hex").slice(0, 16) : null,
    sessionId: trace.sessionId ? crypto.createHash("sha256").update(trace.sessionId).digest("hex").slice(0, 16) : null,
    input: "[REDACTED]",
    output: "[REDACTED]",
    metadata: {
      model: trace.metadata?.model,
      // Keep operational fields, remove PII
    },
  };
}

async function exportAnonymized(fromDate: string, outputFile: string) {
  const traces = await langfuse.api.traces.list({
    fromTimestamp: fromDate,
    limit: 1000,
  });

  const anonymized = traces.data.map(anonymizeTrace);
  writeFileSync(outputFile, JSON.stringify(anonymized, null, 2));
}

Data Categories and Retention

Error Handling

Output

Produce either a redacted, access-controlled export with its date range and trace count, or a deletion receipt containing only the request identifier, affected-user identifier, and deleted-record count. Never place raw prompts, outputs, or API keys in the completion report.

Examples

For a subject-access request, export one user's records to an encrypted controlled location, verify the trace count against the dashboard, and record the export date and retention deadline. For an erasure request, use the exact trace user ID, wait for the documented propagation period, then query again to confirm no matching traces remain.

Resources

• Langfuse Data Security
• Public API Reference
• API Reference (OpenAPI)
• Self-Hosting Configuration