> For clean Markdown of any page, append .md to the page URL. > For a complete documentation index, see https://docs.nvidia.com/nemo-platform/llms.txt. > For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/nemo-platform/_mcp/server. # Create Config POST https://host.com/apis/guardrails/v2/workspaces/{workspace}/configs Content-Type: application/json Create a new guardrail config. Reference: https://docs.nvidia.com/nemo-platform/nemo-platform/documentation/reference/api-reference/guardrails/create-config-apis-guardrails-v-2-workspaces-workspace-configs-post ## OpenAPI Specification ```yaml openapi: 3.1.0 info: title: Nemo Platform API version: 1.0.0 paths: /apis/guardrails/v2/workspaces/{workspace}/configs: post: operationId: create-config-apis-guardrails-v-2-workspaces-workspace-configs-post summary: Create Config description: Create a new guardrail config. tags: - subpackage_guardrails parameters: - name: workspace in: path required: true schema: type: string responses: '201': description: Config created successfully. content: application/json: schema: $ref: '#/components/schemas/GuardrailConfig' '409': description: Config already exists. content: application/json: schema: description: Any type '422': description: Validation Error content: application/json: schema: description: Any type requestBody: content: application/json: schema: $ref: '#/components/schemas/GuardrailConfigInput' servers: - url: https://host.com description: Default components: schemas: GuardrailConfigInput: type: object properties: name: type: string description: The name of the guardrail config description: type: string description: Description of the guardrail config data: type: object additionalProperties: description: Any type description: Guardrail configuration data required: - name description: Input schema for creating a guardrail config. title: GuardrailConfigInput ModelParameters: type: object properties: base_url: type: string description: The URL to use for inference with this model. default_headers: type: object additionalProperties: type: string description: >- Custom HTTP headers to include in requests to this model. Each key-value pair represents a header name (key) and its default value (value). You can override the default value for a header by populating it in the request headers. description: >- Parameters for configuring how to interact with a model in a guardrails config. title: ModelParameters ModelMode: type: string enum: - chat - text default: chat description: >- Whether the mode is 'text' completion or 'chat' completion. Allowed values are 'chat' or 'text'. title: ModelMode CacheStatsConfig: type: object properties: enabled: type: boolean default: false description: Whether cache statistics tracking is enabled log_interval: type: number format: double description: >- Seconds between periodic cache stats logging to logs (None disables logging) description: Configuration for cache statistics tracking and logging. title: CacheStatsConfig ModelCacheConfig: type: object properties: enabled: type: boolean default: false description: 'Whether caching is enabled (default: False - no caching)' maxsize: type: integer default: 50000 description: Maximum number of entries in the cache per model stats: $ref: '#/components/schemas/CacheStatsConfig' description: Configuration for cache statistics tracking and logging description: Configuration for model caching. title: ModelCacheConfig Model: type: object properties: type: type: string engine: type: string model: type: string description: >- The model name. If using Inference Gateway, this should be the Model Entity reference ('workspace/model_name'). parameters: $ref: '#/components/schemas/ModelParameters' description: Additional parameters to configure how to interact with the model. mode: $ref: '#/components/schemas/ModelMode' default: chat description: >- Whether the mode is 'text' completion or 'chat' completion. Allowed values are 'chat' or 'text'. cache: $ref: '#/components/schemas/ModelCacheConfig' description: >- Cache configuration for this specific model (primarily used for content safety models) required: - type - engine description: >- Configuration of a model used by the rails engine. If using Inference Gateway, the `model` field should be a Model Entity reference ('workspace/model_name'). title: Model Instruction: type: object properties: type: type: string content: type: string required: - type - content description: >- Configuration for instructions in natural language that should be passed to the LLM. title: Instruction MessageTemplate: type: object properties: type: type: string description: The type of message, e.g., 'assistant', 'user', 'system'. content: type: string description: The content of the message. required: - type - content description: Template for a message structure. title: MessageTemplate TaskPromptMessagesItems: oneOf: - $ref: '#/components/schemas/MessageTemplate' - type: string title: TaskPromptMessagesItems TaskPrompt: type: object properties: task: type: string description: The id of the task associated with this prompt. content: type: string description: The content of the prompt, if it's a string. messages: type: array items: $ref: '#/components/schemas/TaskPromptMessagesItems' description: The list of messages included in the prompt. Used for chat models. models: type: array items: type: string description: >- If specified, the prompt will be used only for the given LLM engines/models. The format is a list of strings with the format: or /. output_parser: type: string description: The name of the output parser to use for this prompt. max_length: type: integer default: 16000 description: The maximum length of the prompt in number of characters. mode: type: string default: standard description: >- Corresponds to the `prompting_mode` for which this prompt is fetched. Default is 'standard'. stop: type: array items: type: string description: >- If specified, will be configure stop tokens for models that support this. max_tokens: type: integer description: >- The maximum number of tokens that can be generated in the chat completion. required: - task description: Configuration for prompts that will be used for a specific task. title: TaskPrompt FactCheckingRailConfig: type: object properties: parameters: type: object additionalProperties: description: Any type fallback_to_self_check: type: boolean default: false description: Whether to fall back to self-check if another method fail. description: Configuration data for the fact-checking rail. title: FactCheckingRailConfig AutoAlignOptions: type: object properties: guardrails_config: type: object additionalProperties: description: Any type description: >- The guardrails configuration that is passed to the AutoAlign endpoint description: List of guardrails that are activated title: AutoAlignOptions AutoAlignRailConfig: type: object properties: parameters: type: object additionalProperties: description: Any type input: $ref: '#/components/schemas/AutoAlignOptions' description: Input configuration for AutoAlign guardrails output: $ref: '#/components/schemas/AutoAlignOptions' description: Output configuration for AutoAlign guardrails description: Configuration data for the AutoAlign API title: AutoAlignRailConfig PatronusEvaluationSuccessStrategy: type: string enum: - all_pass - any_pass description: |- Strategy for determining whether a Patronus Evaluation API request should pass, especially when multiple evaluators are called in a single request. ALL_PASS requires all evaluators to pass for success. ANY_PASS requires only one evaluator to pass for success. title: PatronusEvaluationSuccessStrategy PatronusEvaluateApiParams: type: object properties: success_strategy: $ref: '#/components/schemas/PatronusEvaluationSuccessStrategy' default: all_pass description: >- Strategy to determine whether the Patronus Evaluate API Guardrail passes or not. params: type: object additionalProperties: description: Any type description: Parameters to the Patronus Evaluate API description: Config to parameterize the Patronus Evaluate API call title: PatronusEvaluateApiParams PatronusEvaluateConfigOutput: type: object properties: evaluate_config: $ref: '#/components/schemas/PatronusEvaluateApiParams' description: Configuration passed to the Patronus Evaluate API description: Config for the Patronus Evaluate API call title: PatronusEvaluateConfigOutput PatronusRailConfigOutput: type: object properties: input: $ref: '#/components/schemas/PatronusEvaluateConfigOutput' description: Patronus Evaluate API configuration for an Input Guardrail output: $ref: '#/components/schemas/PatronusEvaluateConfigOutput' description: Patronus Evaluate API configuration for an Output Guardrail description: Configuration data for the Patronus Evaluate API title: PatronusRailConfigOutput SensitiveDataDetectionOptions: type: object properties: entities: type: array items: type: string description: >- The list of entities that should be detected. Check out https://microsoft.github.io/presidio/supported_entities/ forthe list of supported entities. mask_token: type: string default: '*' description: The token that should be used to mask the sensitive data. score_threshold: type: number format: double default: 0.2 description: >- The score threshold that should be used to detect the sensitive data. title: SensitiveDataDetectionOptions SensitiveDataDetection: type: object properties: recognizers: type: array items: type: object additionalProperties: description: Any type description: >- Additional custom recognizers. Check out https://microsoft.github.io/presidio/tutorial/08_no_code/ for more details. input: $ref: '#/components/schemas/SensitiveDataDetectionOptions' description: Configuration of the entities to be detected on the user input. output: $ref: '#/components/schemas/SensitiveDataDetectionOptions' description: Configuration of the entities to be detected on the bot output. retrieval: $ref: '#/components/schemas/SensitiveDataDetectionOptions' description: >- Configuration of the entities to be detected on retrieved relevant chunks. description: Configuration of what sensitive data should be detected. title: SensitiveDataDetection RegexDetectionOptions: type: object properties: patterns: type: array items: type: string description: List of regex patterns to match against the text. case_insensitive: type: boolean default: false description: Whether to perform case-insensitive matching. description: Configuration options for regex pattern detection on a specific source. title: RegexDetectionOptions RegexDetection: type: object properties: input: $ref: '#/components/schemas/RegexDetectionOptions' description: Configuration for regex patterns to detect on user input. output: $ref: '#/components/schemas/RegexDetectionOptions' description: Configuration for regex patterns to detect on bot output. retrieval: $ref: '#/components/schemas/RegexDetectionOptions' description: >- Configuration for regex patterns to detect on retrieved relevant chunks. description: Configuration for regex pattern detection. title: RegexDetection JailbreakDetectionConfig: type: object properties: server_endpoint: type: string description: The endpoint for the jailbreak detection heuristics/model container. length_per_perplexity_threshold: type: number format: double default: 89.79 description: The length/perplexity threshold. prefix_suffix_perplexity_threshold: type: number format: double default: 1845.65 description: The prefix/suffix perplexity threshold. nim_base_url: type: string description: >- Base URL for jailbreak detection model. Example: http://localhost:8000/v1 nim_server_endpoint: type: string default: classify description: >- Classification path uri. Defaults to 'classify' for NemoGuard JailbreakDetect. api_key: type: string format: password description: >- Secret String with API key for use in Jailbreak requests. Takes precedence over api_key_env_var api_key_env_var: type: string description: >- Environment variable containing API key for jailbreak detection model nim_url: type: string description: 'DEPRECATED: Use nim_base_url instead' nim_port: type: integer description: 'DEPRECATED: Include port in nim_base_url instead' embedding: type: string description: Configuration data for jailbreak detection. title: JailbreakDetectionConfig InjectionDetection: type: object properties: injections: type: array items: type: string description: >- The list of injection types to detect. Options are 'sqli', 'template', 'code', 'xss'.Currently, only SQL injection, template injection, code injection, and markdown cross-site scripting are supported. Custom rules can be added, provided they are in the `yara_path` and have a `.yara` file extension. action: type: string default: reject description: >- Action to take. Options are 'reject' to offer a rejection message, 'omit' to mask the offending content, and 'sanitize' to pass the content as-is in the safest way. These options are listed in descending order of relative safety. 'sanitize' is not implemented at this time. yara_rules: type: object additionalProperties: type: string description: >- Dictionary mapping rule names to YARA rule strings. If provided, these rules will be used instead of loading rules from yara_path. Each rule should be a valid YARA rule string. title: InjectionDetection PrivateAIDetectionOptions: type: object properties: entities: type: array items: type: string description: The list of entities that should be detected. description: Configuration options for Private AI. title: PrivateAIDetectionOptions PrivateAIDetection: type: object properties: server_endpoint: type: string description: The endpoint for the private AI detection server. input: $ref: '#/components/schemas/PrivateAIDetectionOptions' description: Configuration of the entities to be detected on the user input. output: $ref: '#/components/schemas/PrivateAIDetectionOptions' description: Configuration of the entities to be detected on the bot output. retrieval: $ref: '#/components/schemas/PrivateAIDetectionOptions' description: >- Configuration of the entities to be detected on retrieved relevant chunks. description: Configuration for Private AI. title: PrivateAIDetection GLiNERDetectionOptions: type: object properties: entities: type: array items: type: string description: >- The list of entity labels to detect (e.g., 'email', 'phone_number', 'ssn'). description: Configuration options for GLiNER. title: GLiNERDetectionOptions GLiNERDetection: type: object properties: server_endpoint: type: string default: http://localhost:1235/v1/extract description: The endpoint for the GLiNER detection server. threshold: type: number format: double default: 0.5 description: Confidence threshold for entity detection (0.0 to 1.0). chunk_length: type: integer default: 384 description: Length of text chunks for processing. overlap: type: integer default: 128 description: Overlap between chunks. flat_ner: type: boolean default: false description: >- Whether to use flat NER mode. Setting to False allows for nested entities. input: $ref: '#/components/schemas/GLiNERDetectionOptions' description: Configuration of the entities to be detected on the user input. output: $ref: '#/components/schemas/GLiNERDetectionOptions' description: Configuration of the entities to be detected on the bot output. retrieval: $ref: '#/components/schemas/GLiNERDetectionOptions' description: >- Configuration of the entities to be detected on retrieved relevant chunks. description: Configuration for GLiNER PII detection. title: GLiNERDetection FiddlerGuardrails: type: object properties: fiddler_endpoint: type: string default: http://localhost:8080/process/text description: The global endpoint for Fiddler Guardrails requests. safety_threshold: type: number format: double default: 0.1 description: Fiddler Guardrails safety detection threshold. faithfulness_threshold: type: number format: double default: 0.05 description: Fiddler Guardrails faithfulness detection threshold. description: Configuration for Fiddler Guardrails. title: FiddlerGuardrails ClavataRailConfigLabelMatchLogic: type: string enum: - ANY - ALL default: ANY description: |- The logic to use when deciding whether the evaluation matched. If ANY, only one of the configured labels needs to be found in the input or output. If ALL, all configured labels must be found in the input or output. title: ClavataRailConfigLabelMatchLogic ClavataRailOptions: type: object properties: policy: type: string description: The policy alias to use when evaluating inputs or outputs. labels: type: array items: type: string description: |- A list of labels to match against the policy. If no labels are provided, the overall policy result will be returned. If labels are provided, only hits on the provided labels will be considered a hit. required: - policy description: Configuration data for the Clavata API title: ClavataRailOptions ClavataRailConfig: type: object properties: server_endpoint: type: string default: https://gateway.app.clavata.ai:8443 description: The endpoint for the Clavata API policies: type: object additionalProperties: type: string description: A dictionary of policy aliases and their corresponding IDs. label_match_logic: $ref: '#/components/schemas/ClavataRailConfigLabelMatchLogic' default: ANY description: |- The logic to use when deciding whether the evaluation matched. If ANY, only one of the configured labels needs to be found in the input or output. If ALL, all configured labels must be found in the input or output. input: $ref: '#/components/schemas/ClavataRailOptions' description: Clavata configuration for an Input Guardrail output: $ref: '#/components/schemas/ClavataRailOptions' description: Clavata configuration for an Output Guardrail description: Configuration data for the Clavata API title: ClavataRailConfig CrowdStrikeAIDRRailConfig: type: object properties: timeout: type: number format: double default: 30 description: Timeout in seconds for API requests to CrowdStrike AIDR description: Configuration data for the CrowdStrike AIDR API title: CrowdStrikeAIDRRailConfig PangeaRailOptions: type: object properties: recipe: type: string description: >- Recipe key of a configuration of data types and settings defined in the Pangea User Console. It specifies the rules that are to be applied to the text, such as defang malicious URLs. required: - recipe description: Configuration data for the Pangea AI Guard API title: PangeaRailOptions PangeaRailConfig: type: object properties: input: $ref: '#/components/schemas/PangeaRailOptions' description: Pangea configuration for an Input Guardrail output: $ref: '#/components/schemas/PangeaRailOptions' description: Pangea configuration for an Output Guardrail description: Configuration data for the Pangea AI Guard API title: PangeaRailConfig GuardrailsAIValidatorConfig: type: object properties: name: type: string description: >- Unique identifier or import path for the Guardrails AI validator (e.g., 'toxic_language', 'pii', 'regex_match', or 'guardrails/competitor_check'). parameters: type: object additionalProperties: description: Any type description: >- Parameters to pass to the validator during initialization (e.g., threshold, regex pattern). metadata: type: object additionalProperties: description: Any type description: >- Metadata to pass to the validator during validation (e.g., valid_topics, context). required: - name description: Configuration for a single Guardrails AI validator. title: GuardrailsAIValidatorConfig GuardrailsAIRailConfig: type: object properties: validators: type: array items: $ref: '#/components/schemas/GuardrailsAIValidatorConfig' description: >- List of Guardrails AI validators to apply. Each validator can have its own parameters and metadata. description: Configuration data for Guardrails AI integration. title: GuardrailsAIRailConfig TrendMicroRailConfig: type: object properties: v1_url: type: string default: https://api.xdr.trendmicro.com/v3.0/aiSecurity/applyGuardrails description: >- The endpoint for the Trend Micro AI Guard API. For other regions, use: https://api.{region}.xdr.trendmicro.com/v3.0/aiSecurity/applyGuardrails where region is eu, jp, au, in, sg, or mea. api_key_env_var: type: string description: Environment variable containing API key for Trend Micro AI Guard application_name: type: string default: nemo-guardrails description: >- Application name for TMV1-Application-Name header (REQUIRED). Must contain only letters, numbers, hyphens, and underscores, with a maximum length of 64 characters. detailed_response: type: boolean default: false description: >- If True, returns detailed AI Guard results with confidence scores (Prefer: return=representation). If False, returns minimal response with only action and reasons (Prefer: return=minimal). description: Configuration data for the Trend Micro AI Guard API title: TrendMicroRailConfig AIDefenseRailConfig: type: object properties: timeout: type: number format: double default: 30 description: Timeout in seconds for API requests to AI Defense service fail_open: type: boolean default: false description: >- If True, allow content when AI Defense API call fails (fail open). If False, block content when API call fails (fail closed). Does not affect missing configuration validation. description: Configuration data for the Cisco AI Defense API title: AIDefenseRailConfig MultilingualConfig: type: object properties: enabled: type: boolean default: false description: >- If True, detect the language of user input and return refusal messages in the same language. Supported languages: en (English), es (Spanish), zh (Chinese), de (German), fr (French), hi (Hindi), ja (Japanese), ar (Arabic), th (Thai). refusal_messages: type: object additionalProperties: type: string description: >- Custom refusal messages per language code. If not specified, built-in defaults are used. Example: {'en': 'Sorry, I cannot help.', 'es': 'Lo siento, no puedo ayudar.'} description: Configuration for multilingual refusal messages. title: MultilingualConfig ReasoningConfig: type: object properties: enabled: type: boolean default: false description: >- If True, enable reasoning mode (with traces) for content safety models. If False, use low-latency mode without reasoning traces. description: Configuration for reasoning mode in content safety models. title: ReasoningConfig ContentSafetyConfig: type: object properties: multilingual: $ref: '#/components/schemas/MultilingualConfig' reasoning: $ref: '#/components/schemas/ReasoningConfig' description: Configuration data for content safety rails. title: ContentSafetyConfig RailsConfigDataOutput: type: object properties: fact_checking: $ref: '#/components/schemas/FactCheckingRailConfig' description: Configuration data for the fact-checking rail. autoalign: $ref: '#/components/schemas/AutoAlignRailConfig' description: Configuration data for the AutoAlign guardrails API. patronus: $ref: '#/components/schemas/PatronusRailConfigOutput' description: Configuration data for the Patronus Evaluate API. sensitive_data_detection: $ref: '#/components/schemas/SensitiveDataDetection' description: Configuration for detecting sensitive data. regex_detection: $ref: '#/components/schemas/RegexDetection' description: Configuration for regex pattern detection. jailbreak_detection: $ref: '#/components/schemas/JailbreakDetectionConfig' description: Configuration for jailbreak detection. injection_detection: $ref: '#/components/schemas/InjectionDetection' description: Configuration for injection detection. privateai: $ref: '#/components/schemas/PrivateAIDetection' description: Configuration for Private AI. gliner: $ref: '#/components/schemas/GLiNERDetection' description: Configuration for GLiNER PII detection. fiddler: $ref: '#/components/schemas/FiddlerGuardrails' description: Configuration for Fiddler Guardrails. clavata: $ref: '#/components/schemas/ClavataRailConfig' description: Configuration for Clavata. crowdstrike_aidr: $ref: '#/components/schemas/CrowdStrikeAIDRRailConfig' description: Configuration for CrowdStrike AIDR. pangea: $ref: '#/components/schemas/PangeaRailConfig' description: Configuration for Pangea. guardrails_ai: $ref: '#/components/schemas/GuardrailsAIRailConfig' description: Configuration for Guardrails AI validators. trend_micro: $ref: '#/components/schemas/TrendMicroRailConfig' description: Configuration for Trend Micro. ai_defense: $ref: '#/components/schemas/AIDefenseRailConfig' description: Configuration for Cisco AI Defense. content_safety: $ref: '#/components/schemas/ContentSafetyConfig' description: Configuration for content safety rails. description: Configuration data for specific rails that are supported out-of-the-box. title: RailsConfigDataOutput InputRails: type: object properties: parallel: type: boolean default: false description: If True, the input rails are executed in parallel. flows: type: array items: type: string description: The names of all the flows that implement input rails. description: Configuration of input rails. title: InputRails OutputRailsStreamingConfig: type: object properties: enabled: type: boolean default: true description: Enables streaming mode when True. chunk_size: type: integer default: 200 description: >- The number of tokens in each processing chunk. This is the size of the token block on which output rails are applied. context_size: type: integer default: 50 description: >- The number of tokens carried over from the previous chunk to provide context for continuity in processing. stream_first: type: boolean default: true description: >- If True, token chunks are streamed immediately before output rails are applied. description: Configuration for managing streaming output of LLM tokens. title: OutputRailsStreamingConfig OutputRails: type: object properties: parallel: type: boolean default: false description: If True, the output rails are executed in parallel. flows: type: array items: type: string description: The names of all the flows that implement output rails. streaming: $ref: '#/components/schemas/OutputRailsStreamingConfig' description: Configuration for streaming output rails. apply_to_reasoning_traces: type: boolean default: false description: >- If True, output rails will apply guardrails to both reasoning traces and output response. If False, output rails will only apply guardrails to the output response excluding the reasoning traces, thus keeping reasoning traces unaltered. description: Configuration of output rails. title: OutputRails RetrievalRails: type: object properties: flows: type: array items: type: string description: The names of all the flows that implement retrieval rails. description: Configuration of retrieval rails. title: RetrievalRails SingleCallConfig: type: object properties: enabled: type: boolean default: false fallback_to_multiple_calls: type: boolean default: true description: >- Whether to fall back to multiple calls if a single call is not possible. description: Configuration for the single LLM call option for topical rails. title: SingleCallConfig UserMessagesConfig: type: object properties: embeddings_only: type: boolean default: false description: >- Whether to use only embeddings for computing the user canonical form messages. embeddings_only_similarity_threshold: type: number format: double description: >- The similarity threshold to use when using only embeddings for computing the user canonical form messages. embeddings_only_fallback_intent: type: string description: >- Defines the fallback intent when the similarity is below the threshold. If set to None, the user intent is computed normally using the LLM. If set to a string value, that string is used as the intent. description: Configuration for how the user messages are interpreted. title: UserMessagesConfig DialogRails: type: object properties: single_call: $ref: '#/components/schemas/SingleCallConfig' description: Configuration for the single LLM call option. user_messages: $ref: '#/components/schemas/UserMessagesConfig' description: Configuration of topical rails. title: DialogRails ActionRails: type: object properties: instant_actions: type: array items: type: string description: The names of all actions which should finish instantly. description: >- Configuration of action rails. Action rails control various options related to the execution of actions. Currently, only In the future multiple options will be added, e.g., what input validation should be performed per action, output validation, throttling, disabling, etc. title: ActionRails ToolOutputRails: type: object properties: flows: type: array items: type: string description: The names of all the flows that implement tool output rails. parallel: type: boolean default: false description: If True, the tool output rails are executed in parallel. description: >- Configuration of tool output rails. Tool output rails are applied to tool calls before they are executed. They can validate tool names, parameters, and context to ensure safe tool usage. title: ToolOutputRails ToolInputRails: type: object properties: flows: type: array items: type: string description: The names of all the flows that implement tool input rails. parallel: type: boolean default: false description: If True, the tool input rails are executed in parallel. description: >- Configuration of tool input rails. Tool input rails are applied to tool results before they are processed. They can validate, filter, or transform tool outputs for security and safety. title: ToolInputRails RailsOutput: type: object properties: config: $ref: '#/components/schemas/RailsConfigDataOutput' description: >- Configuration data for specific rails that are supported out-of-the-box. input: $ref: '#/components/schemas/InputRails' description: Configuration of the input rails. output: $ref: '#/components/schemas/OutputRails' description: Configuration of the output rails. retrieval: $ref: '#/components/schemas/RetrievalRails' description: Configuration of the retrieval rails. dialog: $ref: '#/components/schemas/DialogRails' description: Configuration of the dialog rails. actions: $ref: '#/components/schemas/ActionRails' description: Configuration of action rails. tool_output: $ref: '#/components/schemas/ToolOutputRails' description: Configuration of tool output rails. tool_input: $ref: '#/components/schemas/ToolInputRails' description: Configuration of tool input rails. description: Configuration of specific rails. title: RailsOutput LogAdapterConfig: type: object properties: name: type: string default: FileSystem description: The name of the adapter. title: LogAdapterConfig TracingConfig: type: object properties: enabled: type: boolean default: false adapters: type: array items: $ref: '#/components/schemas/LogAdapterConfig' description: >- The list of tracing adapters to use. If not specified, the default adapters are used. span_format: type: string default: opentelemetry description: >- The span format to use. Options are 'legacy' (simple metrics) or 'opentelemetry' (OpenTelemetry semantic conventions). enable_content_capture: type: boolean default: false description: >- Capture prompts and responses (user/assistant/tool message content) in tracing/telemetry events. Disabled by default for privacy and alignment with OpenTelemetry GenAI semantic conventions. WARNING: Enabling this may include PII and sensitive data in your telemetry backend. title: TracingConfig RailsConfigOutput: type: object properties: models: type: array items: $ref: '#/components/schemas/Model' description: The list of models used by the rails configuration. instructions: type: array items: $ref: '#/components/schemas/Instruction' default: - type: general content: >- Below is a conversation between a helpful AI assistant and a user. The bot is designed to generate human-like text based on the input that it receives. The bot is talkative and provides lots of specific details. If the bot does not know the answer to a question, it truthfully says it does not know. description: List of instructions in natural language that the LLM should use. actions_server_url: type: string description: The URL of the actions server that should be used for the rails. sample_conversation: type: string default: | user "Hello there!" express greeting bot express greeting "Hello! How can I assist you today?" user "What can you do for me?" ask about capabilities bot respond about capabilities "As an AI assistant, I can help you with a wide range of tasks. This includes question answering on various topics, generating text for various purposes and providing suggestions based on your preferences." user "Tell me a bit about the history of NVIDIA." ask general question bot response for general question "NVIDIA is a technology company that specializes in designing and manufacturing graphics processing units (GPUs) and other computer hardware. The company was founded in 1993 by Jen-Hsun Huang, Chris Malachowsky, and Curtis Priem." user "tell me more" request more information bot provide more information "Initially, the company focused on developing 3D graphics processing technology for the PC gaming market. In 1999, NVIDIA released the GeForce 256, the world's first GPU, which was a major breakthrough for the gaming industry. The company continued to innovate in the GPU space, releasing new products and expanding into other markets such as professional graphics, mobile devices, and artificial intelligence." user "thanks" express appreciation bot express appreciation and offer additional help "You're welcome. If you have any more questions or if there's anything else I can help you with, please don't hesitate to ask." description: The sample conversation that should be used inside the prompts. prompts: type: array items: $ref: '#/components/schemas/TaskPrompt' description: The prompts that should be used for the various LLM tasks. prompting_mode: type: string default: standard description: Allows choosing between different prompting strategies. lowest_temperature: type: number format: double default: 0.001 description: The lowest temperature that should be used for the LLM. enable_multi_step_generation: type: boolean default: false description: Whether to enable multi-step generation for the LLM. colang_version: type: string default: '1.0' description: The Colang version to use. custom_data: type: object additionalProperties: description: Any type description: Any custom configuration data that might be needed. rails: $ref: '#/components/schemas/RailsOutput' description: Configuration for the various rails (input, output, etc.). enable_rails_exceptions: type: boolean default: false description: >- If set, the pre-defined guardrails raise exceptions instead of returning pre-defined messages. passthrough: type: boolean description: >- Whether the original prompt should pass through the guardrails configuration as is. This means it will not be altered in any way. tracing: $ref: '#/components/schemas/TracingConfig' description: Configuration for tracing. description: Configuration object for the models and the rails. title: RailsConfigOutput GuardrailConfig: type: object properties: name: type: string default: '' description: Entity name within the workspace workspace: type: string description: Workspace identifier project: type: string description: The name of the project associated with this entity. description: type: string description: Description of the guardrail config data: $ref: '#/components/schemas/RailsConfigOutput' description: Guardrail configuration data id: type: string created_at: type: string format: date-time created_by: type: - string - 'null' updated_at: type: string format: date-time updated_by: type: - string - 'null' entity_id: type: string description: Alias for id for backwards compatibility. parent: type: string description: Parent entity ID for nested entities. required: - workspace - id - created_at - created_by - updated_at - updated_by - entity_id - parent description: A guardrail configuration entity. title: GuardrailConfig ``` ## Examples **Request** ```json { "name": "string" } ``` **Response** ```json { "workspace": "string", "id": "string", "created_at": "2024-01-15T09:30:00Z", "created_by": "string", "updated_at": "2024-01-15T09:30:00Z", "updated_by": "string", "entity_id": "string", "parent": "string", "name": "", "project": "string", "description": "string", "data": { "models": [ { "type": "string", "engine": "string", "model": "string", "parameters": { "base_url": "string", "default_headers": {} }, "mode": "chat", "cache": { "enabled": false, "maxsize": 50000, "stats": { "enabled": false, "log_interval": 1.1 } } } ], "instructions": [ { "type": "string", "content": "string" } ], "actions_server_url": "string", "sample_conversation": "user \"Hello there!\"\n express greeting\nbot express greeting\n \"Hello! How can I assist you today?\"\nuser \"What can you do for me?\"\n ask about capabilities\nbot respond about capabilities\n \"As an AI assistant, I can help you with a wide range of tasks. This includes question answering on various topics, generating text for various purposes and providing suggestions based on your preferences.\"\nuser \"Tell me a bit about the history of NVIDIA.\"\n ask general question\nbot response for general question\n \"NVIDIA is a technology company that specializes in designing and manufacturing graphics processing units (GPUs) and other computer hardware. The company was founded in 1993 by Jen-Hsun Huang, Chris Malachowsky, and Curtis Priem.\"\nuser \"tell me more\"\n request more information\nbot provide more information\n \"Initially, the company focused on developing 3D graphics processing technology for the PC gaming market. In 1999, NVIDIA released the GeForce 256, the world's first GPU, which was a major breakthrough for the gaming industry. The company continued to innovate in the GPU space, releasing new products and expanding into other markets such as professional graphics, mobile devices, and artificial intelligence.\"\nuser \"thanks\"\n express appreciation\nbot express appreciation and offer additional help\n \"You're welcome. If you have any more questions or if there's anything else I can help you with, please don't hesitate to ask.\"\n", "prompts": [ { "task": "string", "content": "string", "messages": [ { "type": "string", "content": "string" } ], "models": [ "string" ], "output_parser": "string", "max_length": 16000, "mode": "standard", "stop": [ "string" ], "max_tokens": 1 } ], "prompting_mode": "standard", "lowest_temperature": 0.001, "enable_multi_step_generation": false, "colang_version": "1.0", "custom_data": {}, "rails": { "config": { "fact_checking": { "parameters": {}, "fallback_to_self_check": false }, "autoalign": { "parameters": {}, "input": { "guardrails_config": {} }, "output": { "guardrails_config": {} } }, "patronus": { "input": { "evaluate_config": { "success_strategy": "all_pass", "params": {} } }, "output": { "evaluate_config": { "success_strategy": "all_pass", "params": {} } } }, "sensitive_data_detection": { "recognizers": [ {} ], "input": { "entities": [ "string" ], "mask_token": "*", "score_threshold": 0.2 }, "output": { "entities": [ "string" ], "mask_token": "*", "score_threshold": 0.2 }, "retrieval": { "entities": [ "string" ], "mask_token": "*", "score_threshold": 0.2 } }, "regex_detection": { "input": { "patterns": [ "string" ], "case_insensitive": false }, "output": { "patterns": [ "string" ], "case_insensitive": false }, "retrieval": { "patterns": [ "string" ], "case_insensitive": false } }, "jailbreak_detection": { "server_endpoint": "string", "length_per_perplexity_threshold": 89.79, "prefix_suffix_perplexity_threshold": 1845.65, "nim_base_url": "string", "nim_server_endpoint": "classify", "api_key_env_var": "string", "nim_url": "string", "nim_port": 1, "embedding": "string" }, "injection_detection": { "injections": [ "string" ], "action": "reject", "yara_rules": {} }, "privateai": { "server_endpoint": "string", "input": { "entities": [ "string" ] }, "output": { "entities": [ "string" ] }, "retrieval": { "entities": [ "string" ] } }, "gliner": { "server_endpoint": "http://localhost:1235/v1/extract", "threshold": 0.5, "chunk_length": 384, "overlap": 128, "flat_ner": false, "input": { "entities": [ "string" ] }, "output": { "entities": [ "string" ] }, "retrieval": { "entities": [ "string" ] } }, "fiddler": { "fiddler_endpoint": "http://localhost:8080/process/text", "safety_threshold": 0.1, "faithfulness_threshold": 0.05 }, "clavata": { "server_endpoint": "https://gateway.app.clavata.ai:8443", "policies": {}, "label_match_logic": "ANY", "input": { "policy": "string", "labels": [ "string" ] }, "output": { "policy": "string", "labels": [ "string" ] } }, "crowdstrike_aidr": { "timeout": 30 }, "pangea": { "input": { "recipe": "string" }, "output": { "recipe": "string" } }, "guardrails_ai": { "validators": [ { "name": "string", "parameters": {}, "metadata": {} } ] }, "trend_micro": { "v1_url": "https://api.xdr.trendmicro.com/v3.0/aiSecurity/applyGuardrails", "api_key_env_var": "string", "application_name": "nemo-guardrails", "detailed_response": false }, "ai_defense": { "timeout": 30, "fail_open": false }, "content_safety": { "multilingual": { "enabled": false, "refusal_messages": {} }, "reasoning": { "enabled": false } } }, "input": { "parallel": false, "flows": [ "string" ] }, "output": { "parallel": false, "flows": [ "string" ], "streaming": { "enabled": true, "chunk_size": 200, "context_size": 50, "stream_first": true }, "apply_to_reasoning_traces": false }, "retrieval": { "flows": [ "string" ] }, "dialog": { "single_call": { "enabled": false, "fallback_to_multiple_calls": true }, "user_messages": { "embeddings_only": false, "embeddings_only_similarity_threshold": 0.5, "embeddings_only_fallback_intent": "string" } }, "actions": { "instant_actions": [ "string" ] }, "tool_output": { "flows": [ "string" ], "parallel": false }, "tool_input": { "flows": [ "string" ], "parallel": false } }, "enable_rails_exceptions": false, "passthrough": true, "tracing": { "enabled": false, "adapters": [ { "name": "FileSystem" } ], "span_format": "opentelemetry", "enable_content_capture": false } } } ``` **SDK Code** ```python import requests url = "https://host.com/apis/guardrails/v2/workspaces/workspace/configs" payload = { "name": "string" } headers = {"Content-Type": "application/json"} response = requests.post(url, json=payload, headers=headers) print(response.json()) ``` ```javascript const url = 'https://host.com/apis/guardrails/v2/workspaces/workspace/configs'; const options = { method: 'POST', headers: {'Content-Type': 'application/json'}, body: '{"name":"string"}' }; try { const response = await fetch(url, options); const data = await response.json(); console.log(data); } catch (error) { console.error(error); } ``` ```go package main import ( "fmt" "strings" "net/http" "io" ) func main() { url := "https://host.com/apis/guardrails/v2/workspaces/workspace/configs" payload := strings.NewReader("{\n \"name\": \"string\"\n}") req, _ := http.NewRequest("POST", url, payload) req.Header.Add("Content-Type", "application/json") res, _ := http.DefaultClient.Do(req) defer res.Body.Close() body, _ := io.ReadAll(res.Body) fmt.Println(res) fmt.Println(string(body)) } ``` ```ruby require 'uri' require 'net/http' url = URI("https://host.com/apis/guardrails/v2/workspaces/workspace/configs") http = Net::HTTP.new(url.host, url.port) http.use_ssl = true request = Net::HTTP::Post.new(url) request["Content-Type"] = 'application/json' request.body = "{\n \"name\": \"string\"\n}" response = http.request(request) puts response.read_body ``` ```java import com.mashape.unirest.http.HttpResponse; import com.mashape.unirest.http.Unirest; HttpResponse response = Unirest.post("https://host.com/apis/guardrails/v2/workspaces/workspace/configs") .header("Content-Type", "application/json") .body("{\n \"name\": \"string\"\n}") .asString(); ``` ```php request('POST', 'https://host.com/apis/guardrails/v2/workspaces/workspace/configs', [ 'body' => '{ "name": "string" }', 'headers' => [ 'Content-Type' => 'application/json', ], ]); echo $response->getBody(); ``` ```csharp using RestSharp; var client = new RestClient("https://host.com/apis/guardrails/v2/workspaces/workspace/configs"); var request = new RestRequest(Method.POST); request.AddHeader("Content-Type", "application/json"); request.AddParameter("application/json", "{\n \"name\": \"string\"\n}", ParameterType.RequestBody); IRestResponse response = client.Execute(request); ``` ```swift import Foundation let headers = ["Content-Type": "application/json"] let parameters = ["name": "string"] as [String : Any] let postData = JSONSerialization.data(withJSONObject: parameters, options: []) let request = NSMutableURLRequest(url: NSURL(string: "https://host.com/apis/guardrails/v2/workspaces/workspace/configs")! as URL, cachePolicy: .useProtocolCachePolicy, timeoutInterval: 10.0) request.httpMethod = "POST" request.allHTTPHeaderFields = headers request.httpBody = postData as Data let session = URLSession.shared let dataTask = session.dataTask(with: request as URLRequest, completionHandler: { (data, response, error) -> Void in if (error != nil) { print(error as Any) } else { let httpResponse = response as? HTTPURLResponse print(httpResponse) } }) dataTask.resume() ```