Guardrails Hub – Guardrails AI

Guardrails AI has launched the Guardrails Hub, a searchable repository of 70 validators designed to ensure the safety, quality, and compliance of AI-generated text. The hub includes validators for detecting PII, profanity, toxic language, and jailbreak attempts, as well as tools for factuality checks and translation quality, catering to developers integrating guardrails into their LLM applications.

Guardrails AI Hub Blog Docs Loading ... The Guardrails Hub Search and explore vast world of guardrails validators through lightning-fast search 70 validators Validators 70 / 70 Arize Dataset Embeddings Select Validates that user-generated input does not match the dataset of jailbreak embeddings from Arize AI. string Brand risk ML Ban List Select Validates that the output does not contain banned words, using fuzzy search. string Brand risk ML Bespoke MiniCheck Select Validates that the LLM-generated text is supported by the provided context using BespokeLabs.AI's MiniCheck API. string Brand risk +1 more ML Bias Check Select Validates that the text is free from biases related to age, gender, sex, ethnicity, religion, etc. string Brand risk ML Competitor Check Select Flags mentions of competitors. Fixes responses by filtering out competitor names. string Brand risk ML Correct Language Select Validate that an LLM-generated text is in the expected language. If the text is not in the expected language, the validator will attempt to translate it to the expected language. string Etiquette ML Cucumber Expression Match Select Validates that the input string matches a specified cucumber expression. string Brand risk ML Detect PII Select Detects personally identifiable information PII in text, using Microsoft Presidio. string Data Leakage ML Extracted Summary Sentences Match Select This validator checks if the extracted summary sentences match the original document. string Factuality +1 more ML Extractive Summary Select Uses fuzzy matching to detect if some text is a summary of a document. string Factuality Gibberish Text Select A Guardrails AI validator to detect gibberish text. string Brand risk +3 more ML High Quality Translation Select A validator that checks if a translation is of high quality. string Etiquette +1 more ML Llama Guard Select A llama based validator which checks whether a given prompt is safe/unsafe by specifying a set of policies and lists the violating policies when applicable. string Etiquette ML LLM RAG Evaluator Select This validator uses an LLM Judge to decide whether the LLM response is acceptable in a RAG application. string Factuality +1 more LLM Logic Check Select Validates logical consistency and detects logical fallacies in the model output. Attempts to correct logical fallacies if found. string Brand risk ML NSFW Text Select A Guardrails AI validator to detect NSFW text string Etiquette ML Profanity Free Select Checks for profanity in text, using the alt-profanity-check library. string Etiquette +1 more ML Provenance Embeddings Select Compares embeddings of generated and source texts to calculate provenance. string Factuality +1 more ML Provenance LLM Select A validator for ensuring the factuality and reducing brand risk in generated content. string Factuality +2 more ML QA Relevance LLM Eval Select Makes a second request to the LLM, asking it if its original response was relevant to the prompt. string Jailbreaking +1 more LLM Relevancy Evaluator Select Validates that the reference text contains information relevant to answering the original question. string Brand risk ML Restrict to Topic Select Determines if the text pertains to a specified topic. string Etiquette +3 more LLM Saliency Check Select Checks if a generated summary covers topics present in a source document. string Factuality LLM Secrets Present Select Detects the secrets present in text by matching against common patterns for API keys and other sensitive information. string code +2 more Rule Shield Gemma Select A Gemma based validator for moderating user prompts to guard against harmful content by specifying a policy. string Etiquette ML Similar To Document Select Checks if some generated text is similar to a provided document. string Factuality ML Similar To Previous Values Select Checks if a value is similar to a list of previously known correct values. string integer +1 more ML Toxic Language Select Identifies and flags toxic language in text to ensure communications remain professional and appropriate. string Etiquette ML Wiki Provenance Select A Guardrails AI validator that detects and removes hallucinated text based off Wikipedia string Factuality Contains String Select A Guardrails AI validator to check if the LLM-generated text contains a substring. string Formatting CSV Validator Select Checks the CSV file for issues, including mismatched column lengths and inconsistent quote delimiters. csv Formatting +1 more Detect Jailbreak Select Detects attempts to circumvent safeguards in model conditioning. string Brand risk +2 more ML Endpoint Is Reachable Select Checks if an endpoint can be reached by making a request to it. string Code Exploits +1 more Ends With Select Check if a string or list ends with a specified string or list. list string +1 more Exclude SQL Predicates Select This rule checks for the use of particular SQL predicates in the query. It is important to exclude SQL predicates from the query to prevent SQL injection attacks. string sql +3 more Financial Tone Select Validates that an LLM-generated output in a financial context maintains a particular tone. string Etiquette ML Grounded AI Hallucination Select A Grounded AI validator that detects hallucinated text string Factuality Guardrails PII Select Detects personally identifiable information PII in text. string Data Leakage ML Has Url Select Ensure content contains a url. string Code Exploits +1 more LLM Critic Select Grade the generated response based on provided criteria. string Factuality +1 more LLM Lowercase Select Passes when totally lowercase. string Formatting Mentions Drugs Select Validates that the generated text does not contain any drug names string Etiquette ML MLcube RAG Context Evaluator Select A validator that scores retrieved RAG context for relevance and usefulness to the user query. string Factuality One Line Select This validator checks if the input is a single line of text. string Formatting Politeness Check Select Ensure generated output is polite. string Etiquette LLM Prompt Injection Detector Select A Guardrails validator that scores prompts for injection attempts via a secondary LLM. string Jailbreaking LLM Quotes Price Select Validates that the generated text contains a price quote string Brand risk ML Reading Level Select Parses text to find its readability as a US grade level number 0-12 . string Etiquette +1 more Reading Time Select Ensures that any generated text is less than a maximum expected reading time. string Formatting Redundant Sentences Select Identifies redundant sentences in text using fuzzy matching. string Etiquette ML Regex Match Select Ensure content matches a provided regular expression. This can be used to validate content such as email addresses, phone numbers, and more. string Formatting Response Evaluator Select Evaluate generated output using a provided question. string Factuality +2 more LLM Responsiveness Check Select Ensure generated output is polite. string Factuality LLM Sensitive Topic Select A Guardrails AI validator that detects sensitive topics in text. string Etiquette ML Sql Column Presence Select Checks that schema columns are present in a SQL query. string sql +3 more Toxic Language LLM Select Detects toxic language in LLM-generated text using an LLM as the detection backbone. Evaluates text across seven toxicity categories: toxicity, severe toxicity, obscene, threat, insult, identity attack, and sexual explicit content. string Etiquette LLM Two Words Select Passes when value is exactly two words. string Formatting Unusual Prompt Select A Guardrails AI input validator that validates a prompt for unusualness and trickery. string Etiquette +1 more LLM Uppercase Select Passes when totally uppercase. string Formatting Valid Address Select Verifies an LLM-generated address using Google Maps' Address Validation API. string Formatting +1 more Valid Choices Select Checks if a given string is a valid choice from a list of choices. string Formatting Valid HTML Select Guardrails validator that checks for HTML parseability. string Formatting Valid JSON Select Ensure content is parseable as valid JSON. string object +3 more Valid Length Select Ensures the length of a string or list falls between a minimum and maximum. string Formatting Valid OpenAPI Specification Select Ensures that a generated output is a valid OpenAPI Specification. string object +2 more Valid Python Select Validates whether the given Python code is syntactically correct. python Invalid Code Valid Range Select Assess whether a generated number is between a maximum and minimum value. integer float +1 more Valid SQL Select Validates whether the given SQL code is syntactically correct using. Optionally accepts a database schema to validate against using SQLAlchemy. sql Invalid Code +1 more Valid URL Select Validates that text is a syntactically-valid URL string Formatting Web Sanitization Select Scans LLM outputs for strings that could cause browser script execution downstream. string Code Exploits 0 Validators selected Generate Code Filters Use Cases Chatbots Customer Support Structured data RAG Summarization CodeGen Text2SQL Risk Category Etiquette Brand risk Factuality Formatting Invalid Code Jailbreaking Code Exploits Data Leakage Infrastructure Requirements ML LLM NA Rule Content Type string object list integer float sql code csv python Certification Guardrails Certified Language en The Guardrails Hub Search and explore vast world of guardrails validators through lightning-fast search 70 validators Validators 70 / 70 Arize Dataset Embeddings Select Validates that user-generated input does not match the dataset of jailbreak embeddings from Arize AI. string Brand risk ML Ban List Select Validates that the output does not contain banned words, using fuzzy search. string Brand risk ML Bespoke MiniCheck Select Validates that the LLM-generated text is supported by the provided context using BespokeLabs.AI's MiniCheck API. string Brand risk +1 more ML Bias Check Select Validates that the text is free from biases related to age, gender, sex, ethnicity, religion, etc. string Brand risk ML Competitor Check Select Flags mentions of competitors. Fixes responses by filtering out competitor names. string Brand risk ML Correct Language Select Validate that an LLM-generated text is in the expected language. If the text is not in the expected language, the validator will attempt to translate it to the expected language. string Etiquette ML Cucumber Expression Match Select Validates that the input string matches a specified cucumber expression. string Brand risk ML Detect PII Select Detects personally identifiable information PII in text, using Microsoft Presidio. string Data Leakage ML Extracted Summary Sentences Match Select This validator checks if the extracted summary sentences match the original document. string Factuality +1 more ML Extractive Summary Select Uses fuzzy matching to detect if some text is a summary of a document. string Factuality Gibberish Text Select A Guardrails AI validator to detect gibberish text. string Brand risk +3 more ML High Quality Translation Select A validator that checks if a translation is of high quality. string Etiquette +1 more ML Llama Guard Select A llama based validator which checks whether a given prompt is safe/unsafe by specifying a set of policies and lists the violating policies when applicable. string Etiquette ML LLM RAG Evaluator Select This validator uses an LLM Judge to decide whether the LLM response is acceptable in a RAG application. string Factuality +1 more LLM Logic Check Select Validates logical consistency and detects logical fallacies in the model output. Attempts to correct logical fallacies if found. string Brand risk ML NSFW Text Select A Guardrails AI validator to detect NSFW text string Etiquette ML Profanity Free Select Checks for profanity in text, using the alt-profanity-check library. string Etiquette +1 more ML Provenance Embeddings Select Compares embeddings of generated and source texts to calculate provenance. string Factuality +1 more ML Provenance LLM Select A validator for ensuring the factuality and reducing brand risk in generated content. string Factuality +2 more ML QA Relevance LLM Eval Select Makes a second request to the LLM, asking it if its original response was relevant to the prompt. string Jailbreaking +1 more LLM Relevancy Evaluator Select Validates that the reference text contains information relevant to answering the original question. string Brand risk ML Restrict to Topic Select Determines if the text pertains to a specified topic. string Etiquette +3 more LLM Saliency Check Select Checks if a generated summary covers topics present in a source document. string Factuality LLM Secrets Present Select Detects the secrets present in text by matching against common patterns for API keys and other sensitive information. string code +2 more Rule Shield Gemma Select A Gemma based validator for moderating user prompts to guard against harmful content by specifying a policy. string Etiquette ML Similar To Document Select Checks if some generated text is similar to a provided document. string Factuality ML Similar To Previous Values Select Checks if a value is similar to a list of previously known correct values. string integer +1 more ML Toxic Language Select Identifies and flags toxic language in text to ensure communications remain professional and appropriate. string Etiquette ML Wiki Provenance Select A Guardrails AI validator that detects and removes hallucinated text based off Wikipedia string Factuality Contains String Select A Guardrails AI validator to check if the LLM-generated text contains a substring. string Formatting CSV Validator Select Checks the CSV file for issues, including mismatched column lengths and inconsistent quote delimiters. csv Formatting +1 more Detect Jailbreak Select Detects attempts to circumvent safeguards in model conditioning. string Brand risk +2 more ML Endpoint Is Reachable Select Checks if an endpoint can be reached by making a request to it. string Code Exploits +1 more Ends With Select Check if a string or list ends with a specified string or list. list string +1 more Exclude SQL Predicates Select This rule checks for the use of particular SQL predicates in the query. It is important to exclude SQL predicates from the query to prevent SQL injection attacks. string sql +3 more Financial Tone Select Validates that an LLM-generated output in a financial context maintains a particular tone. string Etiquette ML Grounded AI Hallucination Select A Grounded AI validator that detects hallucinated text string Factuality Guardrails PII Select Detects personally identifiable information PII in text. string Data Leakage ML Has Url Select Ensure content contains a url. string Code Exploits +1 more LLM Critic Select Grade the generated response based on provided criteria. string Factuality +1 more LLM Lowercase Select Passes when totally lowercase. string Formatting Mentions Drugs Select Validates that the generated text does not contain any drug names string Etiquette ML MLcube RAG Context Evaluator Select A validator that scores retrieved RAG context for relevance and usefulness to the user query. string Factuality One Line Select This validator checks if the input is a single line of text. string Formatting Politeness Check Select Ensure generated output is polite. string Etiquette LLM Prompt Injection Detector Select A Guardrails validator that scores prompts for injection attempts via a secondary LLM. string Jailbreaking LLM Quotes Price Select Validates that the generated text contains a price quote string Brand risk ML Reading Level Select Parses text to find its readability as a US grade level number 0-12 . string Etiquette +1 more Reading Time Select Ensures that any generated text is less than a maximum expected reading time. string Formatting Redundant Sentences Select Identifies redundant sentences in text using fuzzy matching. string Etiquette ML Regex Match Select Ensure content matches a provided regular expression. This can be used to validate content such as email addresses, phone numbers, and more. string Formatting Response Evaluator Select Evaluate generated output using a provided question. string Factuality +2 more LLM Responsiveness Check Select Ensure generated output is polite. string Factuality LLM Sensitive Topic Select A Guardrails AI validator that detects sensitive topics in text. string Etiquette ML Sql Column Presence Select Checks that schema columns are present in a SQL query. string sql +3 more Toxic Language LLM Select Detects toxic language in LLM-generated text using an LLM as the detection backbone. Evaluates text across seven toxicity categories: toxicity, severe toxicity, obscene, threat, insult, identity attack, and sexual explicit content. string Etiquette LLM Two Words Select Passes when value is exactly two words. string Formatting Unusual Prompt Select A Guardrails AI input validator that validates a prompt for unusualness and trickery. string Etiquette +1 more LLM Uppercase Select Passes when totally uppercase. string Formatting Valid Address Select Verifies an LLM-generated address using Google Maps' Address Validation API. string Formatting +1 more Valid Choices Select Checks if a given string is a valid choice from a list of choices. string Formatting Valid HTML Select Guardrails validator that checks for HTML parseability. string Formatting Valid JSON Select Ensure content is parseable as valid JSON. string object +3 more Valid Length Select Ensures the length of a string or list falls between a minimum and maximum. string Formatting Valid OpenAPI Specification Select Ensures that a generated output is a valid OpenAPI Specification. string object +2 more Valid Python Select Validates whether the given Python code is syntactically correct. python Invalid Code Valid Range Select Assess whether a generated number is between a maximum and minimum value. integer float +1 more Valid SQL Select Validates whether the given SQL code is syntactically correct using. Optionally accepts a database schema to validate against using SQLAlchemy. sql Invalid Code +1 more Valid URL Select Validates that text is a syntactically-valid URL string Formatting Web Sanitization Select Scans LLM outputs for strings that could cause browser script execution downstream. string Code Exploits 0 Validators selected Generate Code