HiddenLayer provides integrations with various third-party products and processes.
Integrations are a key part of your organization’s security and streamline your data flow. From protecting your models, viewing the results, and interacting with HiddenLayer’s API, these integrations are provided to improve your efficiency and enhance your decision-making capabilities.
Connect multiple accounts / workspaces for cloud / data platforms to discover and manage AI-related assets.
| Name | Description | Links |
|---|---|---|
| AWS | AWS AI Asset Discovery automates discovery and inventory of AWS AI Assets, providing a centralized inventory and dashboard. Visibility includes AWS Bedrock, Bedrock Agentcore, and AWS Sagemaker. | AWS Setup Guide |
| Databricks Asset Discovery | Databricks AI Asset Discovery automates discovery and inventory of Databricks AI Assets, providing a centralized inventory and dashboard. Visibility includes discovery into the Unity Catalog. | Databricks Setup Guide |
| Databricks Model Scanner | HiddenLayer integrates with Unity Catalog to automatically scan ML models for vulnerabilities, using the HiddenLayer Supply Chain model scanner. | Databricks Unity Catalog |
| Azure Asset Discovery | Azure AI Asset Discovery automates discovery and inventory of AI Assets, providing a centralized inventory and dashboard. Visibility includes Foundry and Azure ML. | AI Asset Discovery for Azure |
Embed software checks into build pipelines for managing vulnerabilities.
| Name | Description | Links |
|---|---|---|
| Azure DevOps Task | Integrate model scanning into your continuous integration (CI) process with HiddenLayer's Azure DevOps task. | Github |
| Jenkins Plugin | This plugin submits models to the HiddenLayer Model Scanner service for scanning. The plugin can be configured to fail the Jenkins pipeline in the case that the model is flagged as malicious. | Jenkins Plugin |
| JFrog Artifactory | This plugin submits models downloaded from a HuggingFace repository to the HiddenLayer API for scanning. The plugin will block the download if the model is flagged as malicious. | JFrog Artifactory GitHub |
| Model Scanner Github Action | Integrate model scanning into your continuous integration (CI) process with HiddenLayer's GitHub Actions (GHA) integration. This action can scan your models that are stored in a repository or on Amazon S3. | Model Scanner Github |
| Name | Description | Links |
|---|---|---|
| Azure Sentinel | Runtime detections will be sent from the HiddenLayer Console to an Azure Log Analytics Workspace for ingestion by Sentinel. | Azure Sentinel Integration Guide |
| Splunk | The Splunk integration allows HiddenLayer to send Supply Chain, Runtime Security, and Audit Log events as data sources to the specified Splunk endpoint. | Splunk Integration Guide |
| Webhook | A webhook allows communication between applications by allowing a server to push data to a client when events occur. Using a webhook integration allows HiddenLayer to send new detection data to your application. | Webhook Integration Guide |
Drop-in SDK instrumentation or AI gateways for runtime security and policies.
| Name | Description | Links |
|---|---|---|
| AWS Strands Guardrails SDK | HiddenLayer moderation for Strands agents. This package wraps the Strands event loop so every inbound and outbound message is analyzed by HiddenLayer before it reaches a user or tool. Use it to enforce safety policies, redact sensitive data, and block malicious prompts in real time. | Strands Github |
| Azure APIM | Set up an Azure API Management (APIM) instance so that it will route requests to the chat completion to the LLM Proxy while routing any other requests directly to Azure OpenAI. | Configuring Azure API Management |
| Gated Hugging Face | HuggingFace allows providers to secure their models behind gates to limit access. Set up the Gated HuggingFace integration with a shared token to access these repositories. This streamlines the workflow and enables automation through the API. | HuggingFace Gated Repositories Integration |
| LangChain Guardrails | This package provides a LangChain agent middleware that integrates with the HiddenLayer Python SDK to scan, redact, and/or block content before and after the agent executes. | LangChain Guardrails Github |
| LiteLLM | LiteLLM ships with a native integration for HiddenLayer. The proxy sends every request/response to HiddenLayer’s /detection/v1/interactions endpoint so you can block or redact unsafe content before it reaches your users. | LiteLLM HiddenLayer Guardrails |
| Non-Gated HuggingFace | Under the Supply Chain module is the ability to scan HuggingFace models and repositories | Non-Gated Hugging Face |
| OpenAI Guardrails SDK | Drop-in replacement for the Agents SDK Agent that wires HiddenLayer guardrails into agent and tool execution. Agent input/output and every tool call are sent through HiddenLayer's analyze endpoint so prompt-injection and policy violations are caught automatically. | HiddenLayer Guardrails for OpenAI Github |
Build direct integrations with the HiddenLayer SDKs.
| Name | Description | Links |
|---|---|---|
| HiddenLayer Java SDK | The HiddenLayer Java SDK provides convenient access to the HiddenLayer REST API from applications written in Java. | GitHub |
| HiddenLayer Python SDK | The HiddenLayer Python library provides convenient access to the HiddenLayer REST API from any Python 3.9+ application. | GitHub |
| HiddenLayer TypeScript SDK | Provides a set of tools to interact with the HiddenLayer API. This can be used in both Node.js and browser environments. | GitHub |
Integration with AI gateways that unify and secure access to AI models.
| Name | Description | Links |
|---|---|---|
| Azure API Management | Deploy HiddenLayer Runtime Security guardrails to Azure API Management with the hiddenlayer-apim CLI. The integration can evaluate provider-native LLM requests and responses before traffic reaches the model or returns to users. | Azure API Management CLI |
| TrueFoundry AI Gateway | TrueFoundry AI Gateway is the proxy layer that sits between your applications and the LLM providers and MCP Servers. It is an enterprise-grade platform that enables users to access 1000+ LLMs using a unified interface while taking care of observability and governance. |