List of Guardrail Checks
Each Guardrail Check has a specific purpose, it’s own parameters, supported hooks, and sources.
Partner Guardrails
Apora
- Validate Aporia policies
- Define your Aporia policies on your Aporia dashboard and just pass the project ID in Portkey Guardrail check.
Pillar
- Scan Prompts
- Scan Responses For PII, toxicity, prompt injection detection, and more.
Patronus
- Hallucination detection
- Check for conciseness, helpfulness, politeness
- Check for gender, racial bias
- and more!
Pangea
- Text Guard for scanning LLM inputs and outputs
- Analyze and redact text to avoid model manipulation
- Detect malicious content and undesirable data transfers
Acuvity
- Scan Prompts
- Scan Responses
For PII, toxicity, prompt injection detection, and more.
AWS Bedrock
-
Analyze and redact PII to avoid model manipulation
-
Bring your AWS Guardrails directly inside Portkey
and more!
The logic for all of the Guardrail Checks (including Partner Guardrails) is open source.
Bring Your Own Guardrail
We have built Guardrails in a very modular way, and support bringing your own Guardrail using a custom webhook! Learn more here.
Portkey’s Guardrails
Along with the partner Guardrails, there are also deterministic as well as LLM-based Guardrails supported natively by Portkey.
BASIC Guardrails are available on all Portkey plans.
PRO Guardrails are available on Portkey Pro & Enterprise plans.
BASIC — Deterministic Guardrails
Regex Match
Checks if the request or response text matches a regex pattern.
Parameters: rule: string
Supported Hooks: beforeRequestHook, afterRequestHook
Sentence Count
Checks if the content contains a certain number of sentences. Ranges allowed.
Parameters: minSentences: number, maxSentences: number
Supported Hooks: beforeRequestHook, afterRequestHook
Word Count
Checks if the content contains a certain number of words. Ranges allowed.
Parameters: minWords: number, maxWords: number
Supported Hooks: beforeRequestHook, afterRequestHook
Character Count
Checks if the content contains a certain number of characters. Ranges allowed.
Parameters: minCharacters: number, maxCharacters: number
Supported Hooks: beforeRequestHook, afterRequestHook
JSON Schema
Check if the response JSON matches a JSON schema.
Parameters: schema: json
Supported Hooks: afterRequestHook
JSON Keys
Check if the response JSON contains any, all or none of the mentioned keys.
Parameters: keys: array, operator: string
Supported Hooks: afterRequestHook
Contains
Checks if the content contains any, all or none of the words or phrases.
Parameters: words: array, operator: string
Supported Hooks: afterRequestHook
Valid URLs
Checks if all the URLs mentioned in the content are valid
Parameters: onlyDNS: boolean
Supported Hooks: afterRequestHook
Contains Code
Checks if the content contains code of format SQL, Python, TypeScript, etc.
Parameters: format: string
Supported Hooks: afterRequestHook
Lowercase Detection
Check if the given string is lowercase or not.
Parameters: format: string
Supported Hooks: beforeRequestHook, afterRequestHook
Webhook
Makes a webhook request for custom guardrails
Parameters: webhookURL: string, headers: json
Supported Hooks: beforeRequestHook, afterRequestHook
PRO — LLM Guardrails
Moderate Content
Checks if the content passes the mentioned content moderation checks.
Parameters: categories: array
Supported Hooks: beforeRequestHook
Check Language
Checks if the response content is in the mentioned language.
Parameters: language: string
Supported Hooks: beforeRequestHook
Detect PII
Detects Personally Identifiable Information (PII) in the content.
Parameters: categories: array
Supported Hooks: beforeRequestHook, afterRequestHook
Detect Gibberish
Detects if the content is gibberish.
Parameters: boolean
Supported Hooks: beforeRequestHook, afterRequestHook
Contribute Your Guardrail
Integrate your Guardrail platform with Portkey Gateway and reach our growing user base.
Check out some existing integrations to get started.
Was this page helpful?