Apply Deepgram security best practices for API key management and data protection. Use when securing Deepgram integrations, implementing key rotation, or auditing security configurations. Trigger: "deepgram security", "deepgram API key security", "secure deepgram", "deepgram key rotation", "deepgram data protection", "deepgram PII redaction".
Installation
Details
Usage
After installing, this skill will be available to your AI coding assistant.
Verify installation:
npx agent-skills-cli listSkill Instructions
name: deepgram-security-basics description: | Apply Deepgram security best practices for API key management and data protection. Use when securing Deepgram integrations, implementing key rotation, or auditing security configurations. Trigger: "deepgram security", "deepgram API key security", "secure deepgram", "deepgram key rotation", "deepgram data protection", "deepgram PII redaction". allowed-tools: Read, Write, Edit, Grep, Bash(curl:*) version: 1.0.0 license: MIT author: Jeremy Longshore jeremy@intentsolutions.io compatible-with: claude-code, codex, openclaw tags: [saas, deepgram, api, security, compliance]
Deepgram Security Basics
Overview
Security best practices for Deepgram integration: scoped API keys, key rotation, Deepgram's built-in PII redaction feature, client-side temporary keys, SSRF prevention for audio URLs, and audit logging.
Security Checklist
- API keys in environment variables or secret manager (never in code)
- Separate keys per environment (dev/staging/prod)
- Keys scoped to minimum required permissions
- Key rotation schedule (90 days recommended)
- Deepgram
redactoption enabled for PII-sensitive audio - Audio URLs validated (HTTPS only, no private IPs)
- Audit logging on all transcription operations
Instructions
Step 1: Scoped API Keys
Create keys with minimal permissions in Console > Settings > API Keys:
// Production transcription service — only needs listen scope
const sttKey = process.env.DEEPGRAM_STT_KEY; // Scope: listen
// TTS service — only needs speak scope
const ttsKey = process.env.DEEPGRAM_TTS_KEY; // Scope: speak
// Monitoring dashboard — only needs usage read
const monitorKey = process.env.DEEPGRAM_MONITOR_KEY; // Scope: usage:read
// Admin operations — separate key, restricted access
const adminKey = process.env.DEEPGRAM_ADMIN_KEY; // Scope: manage, keys
Step 2: Deepgram Built-in PII Redaction
import { createClient } from '@deepgram/sdk';
const deepgram = createClient(process.env.DEEPGRAM_API_KEY!);
// Deepgram redacts PII directly in the transcript
const { result } = await deepgram.listen.prerecorded.transcribeUrl(
{ url: audioUrl },
{
model: 'nova-3',
smart_format: true,
// Built-in redaction — replaces sensitive data in transcript
redact: ['pci', 'ssn', 'numbers'],
// pci — Credit card numbers → [REDACTED]
// ssn — Social Security numbers → [REDACTED]
// numbers — All numeric sequences → [REDACTED]
}
);
// Transcript will contain [REDACTED] in place of sensitive numbers
console.log(result.results.channels[0].alternatives[0].transcript);
// "My card number is [REDACTED] and my SSN is [REDACTED]"
Step 3: Temporary Keys for Client-Side
// Generate short-lived keys for browser/mobile clients
// This prevents exposing your main API key
import { createClient } from '@deepgram/sdk';
import express from 'express';
const app = express();
const deepgram = createClient(process.env.DEEPGRAM_API_KEY!);
app.post('/api/deepgram/token', async (req, res) => {
// Create a temporary key that expires in 10 seconds
// Use for browser WebSocket connections
const { result, error } = await deepgram.manage.createProjectKey(
process.env.DEEPGRAM_PROJECT_ID!,
{
comment: `temp-key-${Date.now()}`,
scopes: ['listen'], // Minimal scope
time_to_live_in_seconds: 10, // Short-lived
}
);
if (error) return res.status(500).json({ error: error.message });
res.json({ key: result.key, expires_in: 10 });
});
// Browser client uses temporary key:
// const { key } = await fetch('/api/deepgram/token').then(r => r.json());
// const ws = new WebSocket('wss://api.deepgram.com/v1/listen', ['token', key]);
Step 4: Key Rotation
import { createClient } from '@deepgram/sdk';
async function rotateApiKey(projectId: string) {
const admin = createClient(process.env.DEEPGRAM_ADMIN_KEY!);
// 1. Create new key with same scopes
const { result: newKey } = await admin.manage.createProjectKey(projectId, {
comment: `rotated-${new Date().toISOString().split('T')[0]}`,
scopes: ['listen', 'speak'],
expiration_date: new Date(Date.now() + 90 * 86400000).toISOString(), // 90 days
});
console.log('New key created:', newKey.key_id);
// 2. Update secret manager (example: GCP Secret Manager)
// await updateSecret('DEEPGRAM_API_KEY', newKey.key);
// 3. Validate new key works
const testClient = createClient(newKey.key);
const { error } = await testClient.manage.getProjects();
if (error) throw new Error('New key validation failed — aborting rotation');
// 4. Delete old key (after services have picked up new key)
// await admin.manage.deleteProjectKey(projectId, oldKeyId);
return newKey;
}
Step 5: Audio URL Validation (SSRF Prevention)
import { URL } from 'url';
import { lookup } from 'dns/promises';
async function validateAudioUrl(url: string): Promise<void> {
const parsed = new URL(url);
// Require HTTPS
if (parsed.protocol !== 'https:') {
throw new Error('Only HTTPS audio URLs allowed');
}
// Block private/internal IPs
const { address } = await lookup(parsed.hostname);
const privateRanges = [
/^127\./, /^10\./, /^172\.(1[6-9]|2\d|3[01])\./, /^192\.168\./,
/^0\./, /^169\.254\./, /^::1$/, /^fc00:/, /^fe80:/,
];
if (privateRanges.some(r => r.test(address))) {
throw new Error(`Blocked: ${parsed.hostname} resolves to private IP`);
}
// Block known internal hostnames
const blockedHosts = ['localhost', 'metadata.google.internal', '169.254.169.254'];
if (blockedHosts.includes(parsed.hostname)) {
throw new Error(`Blocked hostname: ${parsed.hostname}`);
}
}
// Use before transcription:
await validateAudioUrl(userProvidedUrl);
const { result } = await deepgram.listen.prerecorded.transcribeUrl(
{ url: userProvidedUrl }, { model: 'nova-3' }
);
Step 6: Audit Logging
interface AuditEntry {
timestamp: string;
action: 'transcribe' | 'tts' | 'key_create' | 'key_delete';
userId: string;
requestId?: string;
model: string;
audioDuration?: number;
success: boolean;
error?: string;
ip?: string;
}
function logAudit(entry: AuditEntry) {
// Structured JSON for log aggregation (Datadog, CloudWatch, etc.)
const log = {
...entry,
service: 'deepgram-integration',
level: entry.success ? 'info' : 'error',
};
console.log(JSON.stringify(log));
}
// Usage in transcription middleware
async function transcribeWithAudit(userId: string, url: string, ip: string) {
const start = Date.now();
try {
const { result, error } = await deepgram.listen.prerecorded.transcribeUrl(
{ url }, { model: 'nova-3', smart_format: true }
);
logAudit({
timestamp: new Date().toISOString(),
action: 'transcribe',
userId, model: 'nova-3', ip,
requestId: result?.metadata?.request_id,
audioDuration: result?.metadata?.duration,
success: !error,
error: error?.message,
});
if (error) throw error;
return result;
} catch (err: any) {
logAudit({
timestamp: new Date().toISOString(),
action: 'transcribe',
userId, model: 'nova-3', ip,
success: false, error: err.message,
});
throw err;
}
}
Output
- Scoped API keys per service/environment
- Built-in PII redaction via
redactparameter - Temporary keys for client-side (browser/mobile)
- Key rotation with validation and cleanup
- SSRF-safe audio URL validation
- Structured audit logging
Error Handling
| Issue | Cause | Solution |
|---|---|---|
| 403 after scoping | Key missing required scope | Add scope in Console (e.g., listen) |
| Temp key expired | TTL too short | Increase time_to_live_in_seconds |
| Rotation broke service | New key not propagated | Use overlap period — both keys active |
| Redaction missed PII | Wrong redact option | Use redact: ['pci', 'ssn', 'numbers'] |
Resources
More by jeremylongshore
View allGenerate production-ready Google Cloud code examples from official repositories including ADK samples, Genkit templates, Vertex AI notebooks, and Gemini patterns. Use when asked to "show ADK example" or "provide GCP starter kit". Trigger with relevant phrases based on skill purpose.
Build production Firebase Genkit applications including RAG systems, multi-step flows, and tool calling for Node.js/Python/Go. Deploy to Firebase Functions or Cloud Run with AI monitoring. Use when asked to "create genkit flow" or "implement RAG". Trigger with relevant phrases based on skill purpose.
Build and deploy production-ready generative AI agents using Vertex AI, Gemini models, and Google Cloud infrastructure with RAG, function calling, and multi-modal capabilities. Use when appropriate context detected. Trigger with relevant phrases based on skill purpose.
Validate production readiness of Vertex AI Agent Engine deployments across security, monitoring, performance, compliance, and best practices. Generates weighted scores (0-100%) with actionable remediation plans. Use when asked to validate a deployment, run a production readiness check, audit security posture, or verify compliance for Vertex AI agents. Trigger with "validate deployment", "production readiness", "security audit", "compliance check", "is this agent ready for prod", "check my ADK agent", "review before deploy", or "production readiness check". Make sure to use this skill whenever validating ADK agents for Agent Engine.
