Wardstone

Wardstone is an LLM firewall and AI guardrail API that protects AI applications from prompt attacks, harmful content, data leakage, and suspicious links in a single inference call with ~30ms latency.

It provides bidirectional protection: scan both user inputs and model outputs before they reach your LLM or your users. One API call returns risk bands and confidence scores across all threat categories.

Key capabilities:

• Prompt attack detection: blocks jailbreaks, prompt injections, and system instruction bypasses
• Content moderation: filters violence, hate speech, sexual content, self-harm, and criminal activity
• Data leakage prevention: catches PII like SSNs, credit cards, phone numbers, emails, and addresses
• Unknown link detection: flags suspicious or unverified URLs

Works with any LLM provider (OpenAI, Anthropic, Google, Meta, Mistral, etc.). Drop-in REST API with official SDKs for TypeScript, Python, Go, Ruby, PHP, Java, and C#. No specialized hardware required.

Free tier includes 10,000 API calls/month with no credit card required.