What does the generated JSON config contain?

The config includes an array of blocked words/phrases, an array of regex patterns, a list of enabled PII categories with their detection patterns, and format constraints like maximum length and required JSON structure.

How can I use the generated filter config in my application?

The JSON config is designed to be loaded as a runtime configuration object. Parse it server-side and apply each rule to LLM output before returning it to users. You can implement the matching logic yourself or use it as configuration for moderation middleware.

Can I test the filter before deploying?

Yes — use the built-in test panel to paste sample LLM output and see which rules trigger. Matches are highlighted inline so you can tune your rules before integrating them.

AI Output Filter Builder

Build content filter rules for LLM output: blocked words, regex patterns, PII detection, and format constraints.

Design a complete content filter configuration for LLM output. Add blocked words/phrases, custom regex patterns, enable PII category detection (email, phone, SSN, credit card, IP), and set format constraints (max length, must-contain-JSON). Generates a JSON config and includes a live test panel to validate your rules against sample output.

Blocked words / phrases (one per line)

Regex patterns (one per line)

Format constraints

Max lengthchars

Must contain valid JSON

Generated JSON config

{
  "blockedWords": [],
  "regexPatterns": [],
  "pii": [],
  "formatConstraints": {}
}

Test panel — paste sample output

Related Tools

AICAI Prompt Injection CheckerNEW

Detect prompt injection attacks in text with pattern matching and a 0-10 risk score.

AJDAI Jailbreak Pattern DetectorNEW

Detect DAN, developer mode, roleplay exploits, and encoding tricks in AI prompts.

APDAI PII DetectorNEW

Detect personal information (email, phone, SSN, credit card, IP, date of birth) in text before sending to LLMs.

AISAI Input SanitizerNEW

Remove invisible Unicode, escape injection keywords, and strip dangerous content from LLM input.

Learn More

guide:prompt injection prevention guide:pii handling

FAQ

What does the generated JSON config contain?: The config includes an array of blocked words/phrases, an array of regex patterns, a list of enabled PII categories with their detection patterns, and format constraints like maximum length and required JSON structure.
How can I use the generated filter config in my application?: The JSON config is designed to be loaded as a runtime configuration object. Parse it server-side and apply each rule to LLM output before returning it to users. You can implement the matching logic yourself or use it as configuration for moderation middleware.
Can I test the filter before deploying?: Yes — use the built-in test panel to paste sample LLM output and see which rules trigger. Matches are highlighted inline so you can tune your rules before integrating them.