How do I secure Perplexity applications?

Integrate Wardstone Guard with your Perplexity application to detect prompt injections, jailbreaks, harmful content, and data leakage. Wardstone validates inputs before they reach Perplexity and screens outputs before returning them to users.

What security risks does Perplexity have?

Web-Based Prompt Injection: Malicious websites can craft content specifically to inject instructions into Perplexity's responses. SEO Poisoning Attacks: Attackers can manipulate search rankings to surface malicious content. Citation Laundering: Harmful content can appear legitimate when backed by manipulated citations.

Perplexity + Wardstone

Secure Search-Augmented AI

Secure your Perplexity API applications with Wardstone Guard. Protect search-augmented AI from web-based prompt injection and misinformation.

7 Supported Models· all protected with sub-30ms latency

Sonar Reasoning ProSonar ReasoningSonar ProSonarSonar Deep ResearchR1-1776+1 more

Try the Playground Visit Perplexity

Why Secure Perplexity?

Web-Based Prompt Injection

High Risk

Malicious websites can craft content specifically to inject instructions into Perplexity's responses.

SEO Poisoning Attacks

High Risk

Attackers can manipulate search rankings to surface malicious content.

Citation Laundering

Medium Risk

Harmful content can appear legitimate when backed by manipulated citations.

Security Considerations

1
Web search introduces massive indirect prompt injection risk
2
Search results can contain intentionally misleading content
3
Citations don't guarantee content safety
4
Wardstone screens both queries and retrieved web content

How to Integrate

Add Wardstone to your Perplexity integration
Install Wardstone SDK alongside the Perplexity API client.
Validate search queries
Screen user queries for prompt injection attempts before search.
Screen search-augmented responses
Validate outputs for web-injected harmful content.
Monitor citation sources
Flag responses that cite known-malicious or suspicious domains.

Pricing Note

Perplexity API includes search costs. Wardstone protects against the security risks that search augmentation introduces.

Secure Perplexity with Wardstone

# Step 1: Check user input with Wardstone
curl -X POST "https://api.wardstone.ai/v1/detect" \
  -H "Authorization: Bearer YOUR_WARDSTONE_KEY" \
  -H "Content-Type: application/json" \
  -d '{"text": "User message here"}'
 
# Response: { "prompt_attack": { "detected": false, ... } }
 
# Step 2: If safe, send to Perplexity
curl -X POST "https://api.perplexity.ai/chat/completions" \
  -H "Authorization: Bearer YOUR_PERPLEXITY_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "sonar-pro",
    "messages": [{"role": "user", "content": "User message here"}]
  }'
 
# Step 3: Check Perplexity response with Wardstone before returning to user

Common Use Cases

Research assistants

Fact-checking applications

News aggregation

Market research tools

Knowledge discovery

All Supported Perplexity Models

Wardstone Guard protects all Perplexity models with the same comprehensive security coverage. Whether you're using the latest releases or legacy models still in production, every API call is protected.

Sonar Reasoning Pro

Sonar Reasoning

Sonar Pro

Sonar

Sonar Deep Research

R1-1776

Sonar Huge (coming soon)

Major Provider