# NeuralTrust Docs

## Docs

- [NeuralTrust Docs](https://docs.neuraltrust.ai/index.md)
- [Data privacy](https://docs.neuraltrust.ai/neuraltrust/data-privacy/overview.md): Data sovereignty, GDPR / HIPAA / SOX compliance, and the privacy-by-design architecture that keeps sensitive AI data inside your environment.
- [AWS — Hybrid deployment](https://docs.neuraltrust.ai/neuraltrust/deployment/aws/hybrid.md): Complete walkthrough for deploying the NeuralTrust Data Plane on Amazon EKS with the Control Plane running on NeuralTrust SaaS. Covers cluster prep, full values overlay, ALB ingress, ACM certificates, enrollment, and verification.
- [AWS (EKS) overview](https://docs.neuraltrust.ai/neuraltrust/deployment/aws/overview.md): Install the NeuralTrust Platform on Amazon EKS. Covers AWS Load Balancer Controller, ACM certificates, EBS / EFS storage, IRSA for IAM, and GPU node groups for Firewall workers.
- [AWS — Self-hosted deployment](https://docs.neuraltrust.ai/neuraltrust/deployment/aws/self-hosted.md): Complete walkthrough for deploying the full NeuralTrust Platform (Control Plane + Data Plane + TrustGate + Firewall) on Amazon EKS. Covers cluster prep, full values overlay, ALB ingress for every component, ACM, and verification.
- [Azure — Hybrid deployment](https://docs.neuraltrust.ai/neuraltrust/deployment/azure/hybrid.md): Complete walkthrough for deploying the NeuralTrust Data Plane on Azure Kubernetes Service with the Control Plane running on NeuralTrust SaaS. Covers cluster prep, full values overlay, AGIC or NGINX ingress, Key Vault certificates, enrollment, and verification.
- [Azure (AKS) overview](https://docs.neuraltrust.ai/neuraltrust/deployment/azure/overview.md): Install the NeuralTrust Platform on Azure Kubernetes Service. Covers AGIC or NGINX ingress, Key Vault certificates, Azure Disk / Azure Files storage, Workload Identity, and GPU node pools.
- [Azure — Self-hosted deployment](https://docs.neuraltrust.ai/neuraltrust/deployment/azure/self-hosted.md): Complete walkthrough for deploying the full NeuralTrust Platform (Control Plane + Data Plane + TrustGate + Firewall) on Azure Kubernetes Service. Covers cluster prep, full values overlay, AGIC or NGINX ingress for every component, Key Vault certs, and verification.
- [Configuration scenarios](https://docs.neuraltrust.ai/neuraltrust/deployment/configuration.md): Side-by-side reference for the values files the NeuralTrust Platform Helm chart ships, and ready-made overrides for common deployment topologies.
- [Deployment models](https://docs.neuraltrust.ai/neuraltrust/deployment/deployment-models.md): NeuralTrust Platform supports two main deployment topologies — hybrid (Data Plane in your environment, Control Plane in NeuralTrust SaaS) and self-hosted (everything in your cluster). This page explains the differences, trade-offs, and how to choose.
- [Feature flags reference](https://docs.neuraltrust.ai/neuraltrust/deployment/feature-flags.md): Every toggle that changes what the NeuralTrust Platform Helm chart deploys — PostgreSQL, Redis, Kafka, ClickHouse, image registry, storage class, and secrets. Includes the exact values keys, required databases/users for external services, and the secrets you must pre-create per scenario.
- [Firewall deployment](https://docs.neuraltrust.ai/neuraltrust/deployment/firewall.md): Deploy the NeuralTrust Firewall — gateway plus specialized workers for prompt and response safety, on CPU or GPU.
- [GCP — Hybrid deployment](https://docs.neuraltrust.ai/neuraltrust/deployment/gcp/hybrid.md): Complete walkthrough for deploying the NeuralTrust Data Plane on Google Kubernetes Engine with the Control Plane running on NeuralTrust SaaS. Covers cluster prep, full values overlay, ingress, certificates, GPU workers, enrollment, and verification.
- [GCP (GKE) overview](https://docs.neuraltrust.ai/neuraltrust/deployment/gcp/overview.md): Install the NeuralTrust Platform on Google Kubernetes Engine. GCP is the chart's default target — covers GCE Ingress, Managed Certificates, Persistent Disk storage, Workload Identity, and GPU node pools.
- [GCP — Self-hosted deployment](https://docs.neuraltrust.ai/neuraltrust/deployment/gcp/self-hosted.md): Complete walkthrough for deploying the full NeuralTrust Platform (Control Plane + Data Plane + TrustGate + Firewall) on Google Kubernetes Engine. Covers cluster prep, full values overlay, ingress for all components, Managed Certificates, and verification.
- [Image catalog](https://docs.neuraltrust.ai/neuraltrust/deployment/images.md): Complete inventory of container images the NeuralTrust Platform Helm chart deploys, grouped by subchart and by deployment model. Use this when mirroring images, planning air-gapped installs, or sizing image-pull bandwidth.
- [Vanilla Kubernetes — Hybrid deployment](https://docs.neuraltrust.ai/neuraltrust/deployment/kubernetes/hybrid.md): Complete walkthrough for deploying the NeuralTrust Data Plane on any conformant Kubernetes cluster with the Control Plane running on NeuralTrust SaaS. Covers prerequisites, full values overlay, NGINX or Traefik ingress, cert-manager TLS, enrollment, and verification.
- [Vanilla Kubernetes overview](https://docs.neuraltrust.ai/neuraltrust/deployment/kubernetes/overview.md): Install the NeuralTrust Platform on any conformant Kubernetes cluster — on-prem, bare metal, air-gapped, k3s, RKE2, OKE, IBM Cloud Kubernetes Service, or any other distribution. Covers prerequisites, ingress, storage, and cert-manager.
- [Vanilla Kubernetes — Self-hosted deployment](https://docs.neuraltrust.ai/neuraltrust/deployment/kubernetes/self-hosted.md): Complete walkthrough for deploying the full NeuralTrust Platform (Control Plane + Data Plane + TrustGate + Firewall) on any conformant Kubernetes cluster. Covers prerequisites, full values overlay, ingress for every component, cert-manager TLS, air-gapped installs, and verification.
- [OpenShift — Hybrid deployment](https://docs.neuraltrust.ai/neuraltrust/deployment/openshift/hybrid.md): Complete walkthrough for deploying the NeuralTrust Data Plane on Red Hat OpenShift with the Control Plane on NeuralTrust SaaS. Covers project setup, SCC, image pull secret linking, full values overlay, Routes, enrollment, and verification.
- [OpenShift overview](https://docs.neuraltrust.ai/neuraltrust/deployment/openshift/overview.md): Install the NeuralTrust Platform on Red Hat OpenShift 4.10+. Covers Routes (default) vs Ingress, Security Context Constraints (SCC), `oc` workflow, ROSA / ARO / on-prem, image pull secret linking, and GPU workers.
- [OpenShift — Self-hosted deployment](https://docs.neuraltrust.ai/neuraltrust/deployment/openshift/self-hosted.md): Complete walkthrough for deploying the full NeuralTrust Platform (Control Plane + Data Plane + TrustGate + Firewall) on Red Hat OpenShift 4.10+. Covers project setup, SCC, image pull secret linking, full values overlay, Routes for every component, and verification.
- [Deployment overview](https://docs.neuraltrust.ai/neuraltrust/deployment/overview.md): Where the NeuralTrust Platform runs, how the architecture is laid out, and how to pick the right combination of cloud and deployment model for your environment.
- [Secrets management](https://docs.neuraltrust.ai/neuraltrust/deployment/secrets.md): How NeuralTrust Platform secrets are created, stored, rotated, and managed — including auto-generation, pre-generated secrets, and integration with Vault, Sealed Secrets, and External Secrets Operator.
- [Security posture](https://docs.neuraltrust.ai/neuraltrust/security/overview.md): NeuralTrust's platform-wide security model — authentication, access control, networking, encryption, and compliance guarantees.
- [Advanced](https://docs.neuraltrust.ai/platform/advanced.md): Provision a hybrid data plane in your own AWS, GCP, or Azure account — or connect one you already run — directly from Team Settings.
- [Integrations](https://docs.neuraltrust.ai/platform/alert-integrations.md): Forward NeuralTrust alert findings to your SIEM. Connect Microsoft Sentinel, Datadog, Splunk, Elastic, IBM QRadar, or a generic webhook and stream each new alert as an OCSF Detection Finding.
- [Alerts](https://docs.neuraltrust.ai/platform/alerts.md): Turn TrustGuard and TrustGate telemetry into prioritized, deduplicated alerts. Enable predefined detection use cases, author custom rules, correlate across products, and forward findings to your SIEM.
- [Audit Logs](https://docs.neuraltrust.ai/platform/audit-logs.md): View and monitor security audit logs in NeuralTrust. Track authentication events, user management, and SSO activities. Integrate with your SIEM platform.
- [Break the Glass](https://docs.neuraltrust.ai/platform/break-glass.md): Configure emergency access for administrators to bypass SSO enforcement. Ensure your team is never locked out during identity provider outages.
- [Custom Domain](https://docs.neuraltrust.ai/platform/custom-domain.md): Serve the NeuralTrust app on a domain you own via a CNAME — so users reach the platform at a URL on your organization's zone.
- [Feature Flags](https://docs.neuraltrust.ai/platform/feature-flags.md): Toggle preview capabilities and early-access features for your team — opt in to new functionality before it becomes default.
- [General](https://docs.neuraltrust.ai/platform/general.md): Team identity basics — the team's display name and team deletion.
- [Generic OIDC SSO](https://docs.neuraltrust.ai/platform/generic-oidc-sso.md): Configure Single Sign-On with any OpenID Connect compliant identity provider including Okta, Auth0, Google Workspace, and more.
- [Models](https://docs.neuraltrust.ai/platform/models.md): Choose which LLM and embeddings provider NeuralTrust uses internally — for judge calls, analyzers, semantic classification, and vector operations.
- [Overview](https://docs.neuraltrust.ai/platform/overview.md): Team Settings is where you manage the tenant that hosts your NeuralTrust products — team identity, identity and access, audit, infrastructure, and platform-wide configuration.
- [Password Policy](https://docs.neuraltrust.ai/platform/password-policy.md): Configure password requirements for team members — minimum length, character classes, and blocks for common passwords and personal information.
- [SCIM Provisioning](https://docs.neuraltrust.ai/platform/scim.md): Set up SCIM automatic user provisioning with Microsoft Entra ID. Automatically sync users between Azure AD and NeuralTrust.
- [Microsoft Entra ID SSO](https://docs.neuraltrust.ai/platform/sso.md): Step-by-step guide to configure Microsoft Entra ID (Azure AD) Single Sign-On for NeuralTrust. Enable corporate authentication for your team.
- [Manual User Sync](https://docs.neuraltrust.ai/platform/user-sync.md): Synchronize users from Microsoft Entra ID groups on-demand. Import users with role assignments based on group mappings.
- [Users](https://docs.neuraltrust.ai/platform/users.md): Manage the humans in your NeuralTrust team — members, pending invitations, team roles, and per-product access.
- [Support](https://docs.neuraltrust.ai/support.md): Report bugs and issues, ask the community, or check service status.
- [Create an auth](https://docs.neuraltrust.ai/trustgate/api-reference/auths/create-an-auth.md): Creates a new auth in a gateway.
- [Delete an auth](https://docs.neuraltrust.ai/trustgate/api-reference/auths/delete-an-auth.md): Deletes an auth from a gateway.
- [Get an auth](https://docs.neuraltrust.ai/trustgate/api-reference/auths/get-an-auth.md): Returns a single auth by id.
- [List auths](https://docs.neuraltrust.ai/trustgate/api-reference/auths/list-auths.md): Returns a paginated list of auths in a gateway.
- [Update an auth](https://docs.neuraltrust.ai/trustgate/api-reference/auths/update-an-auth.md): Updates an existing auth.
- [List model catalog](https://docs.neuraltrust.ai/trustgate/api-reference/catalog/list-model-catalog.md): Returns the catalog of supported models, optionally filtered by provider.
- [List policy catalog](https://docs.neuraltrust.ai/trustgate/api-reference/catalog/list-policy-catalog.md): Returns the catalog of available policies grouped by type. Each entry includes the settings schema needed to render its configuration form dynamically.
- [List provider catalog](https://docs.neuraltrust.ai/trustgate/api-reference/catalog/list-provider-catalog.md): Returns the catalog of supported LLM providers.
- [List the MCP servers catalog](https://docs.neuraltrust.ai/trustgate/api-reference/catalog/list-the-mcp-servers-catalog.md): Returns the curated catalog of well-known remote MCP servers, used to prefill MCP registry creation.
- [Attach a policy to a consumer](https://docs.neuraltrust.ai/trustgate/api-reference/consumers/attach-a-policy-to-a-consumer.md): Associates a policy with a consumer (idempotent). Editing the policy later affects every consumer it is attached to.
- [Attach a registry to a consumer](https://docs.neuraltrust.ai/trustgate/api-reference/consumers/attach-a-registry-to-a-consumer.md): Associates a registry with a consumer (idempotent). The optional body sets the registry weight for weighted load balancing.
- [Attach a role to a consumer](https://docs.neuraltrust.ai/trustgate/api-reference/consumers/attach-a-role-to-a-consumer.md): Associates a role with a role_based consumer (idempotent). Returns 409 for inline consumers.
- [Attach an auth to a consumer](https://docs.neuraltrust.ai/trustgate/api-reference/consumers/attach-an-auth-to-a-consumer.md): Associates an auth credential with a consumer (idempotent).
- [Create a consumer](https://docs.neuraltrust.ai/trustgate/api-reference/consumers/create-a-consumer.md): Creates a new consumer in a gateway.
- [Delete a consumer](https://docs.neuraltrust.ai/trustgate/api-reference/consumers/delete-a-consumer.md): Deletes a consumer from a gateway.
- [Detach a policy from a consumer](https://docs.neuraltrust.ai/trustgate/api-reference/consumers/detach-a-policy-from-a-consumer.md): Removes the association between a policy and a consumer (idempotent).
- [Detach a registry from a consumer](https://docs.neuraltrust.ai/trustgate/api-reference/consumers/detach-a-registry-from-a-consumer.md): Removes the association between a registry and a consumer (idempotent).
- [Detach a role from a consumer](https://docs.neuraltrust.ai/trustgate/api-reference/consumers/detach-a-role-from-a-consumer.md): Removes the association between a role and a consumer (idempotent).
- [Detach an auth from a consumer](https://docs.neuraltrust.ai/trustgate/api-reference/consumers/detach-an-auth-from-a-consumer.md): Removes the association between an auth credential and a consumer (idempotent).
- [Get a consumer](https://docs.neuraltrust.ai/trustgate/api-reference/consumers/get-a-consumer.md): Returns a single consumer by id.
- [List consumers](https://docs.neuraltrust.ai/trustgate/api-reference/consumers/list-consumers.md): Returns a paginated list of consumers in a gateway.
- [Update a consumer](https://docs.neuraltrust.ai/trustgate/api-reference/consumers/update-a-consumer.md): Updates an existing consumer.
- [Create a gateway](https://docs.neuraltrust.ai/trustgate/api-reference/gateways/create-a-gateway.md): Creates a new gateway.
- [Delete a gateway](https://docs.neuraltrust.ai/trustgate/api-reference/gateways/delete-a-gateway.md): Deletes a gateway and cascades the deletion to every resource that belongs to it (consumers, roles, policies, auths, registries and vault credentials).
- [Get a gateway](https://docs.neuraltrust.ai/trustgate/api-reference/gateways/get-a-gateway.md): Returns a single gateway by id.
- [List gateways](https://docs.neuraltrust.ai/trustgate/api-reference/gateways/list-gateways.md): Returns a paginated list of gateways.
- [Update a gateway](https://docs.neuraltrust.ai/trustgate/api-reference/gateways/update-a-gateway.md): Updates an existing gateway.
- [Get a playground trace](https://docs.neuraltrust.ai/trustgate/api-reference/playground/get-a-playground-trace.md): Returns the metrics Event captured for a playground request, keyed by the X-AG-Trace-Id returned in the proxy response. Traces expire after a short TTL.
- [Clear a policy's global scope](https://docs.neuraltrust.ai/trustgate/api-reference/policies/clear-a-policys-global-scope.md): Demotes a global policy back to consumer-scoped (applies only to linked consumers).
- [Create a policy](https://docs.neuraltrust.ai/trustgate/api-reference/policies/create-a-policy.md): Creates a new policy in a gateway.
- [Delete a policy](https://docs.neuraltrust.ai/trustgate/api-reference/policies/delete-a-policy.md): Deletes a policy from a gateway.
- [Duplicate a policy](https://docs.neuraltrust.ai/trustgate/api-reference/policies/duplicate-a-policy.md): Creates a copy of an existing policy. The new policy reuses the plugin configuration (slug, settings, stages, enabled, priority, parallel) with a fresh id and an auto-generated name (suffix 2, 3, 4...). The copy has no consumer associations and is not global.
- [Get a policy](https://docs.neuraltrust.ai/trustgate/api-reference/policies/get-a-policy.md): Returns a single policy by id.
- [List policies](https://docs.neuraltrust.ai/trustgate/api-reference/policies/list-policies.md): Returns a paginated list of policies in a gateway.
- [Mark a policy as global](https://docs.neuraltrust.ai/trustgate/api-reference/policies/mark-a-policy-as-global.md): Promotes a policy to gateway-wide scope (applies to every consumer).
- [Update a policy](https://docs.neuraltrust.ai/trustgate/api-reference/policies/update-a-policy.md): Updates an existing policy.
- [Proxy chat completion](https://docs.neuraltrust.ai/trustgate/api-reference/proxy/proxy-chat-completion.md): Forwards an OpenAI Chat Completions request to the selected provider. Proxy plane route: /{consumer_slug}/v1/chat/completions. Other fixed routes include /v1/messages (Anthropic) and /v1/responses (OpenAI Responses).
- [Create a backend](https://docs.neuraltrust.ai/trustgate/api-reference/registries/create-a-backend.md): Creates a new backend in a gateway.
- [Delete a backend](https://docs.neuraltrust.ai/trustgate/api-reference/registries/delete-a-backend.md): Deletes a backend from a gateway.
- [Get a backend](https://docs.neuraltrust.ai/trustgate/api-reference/registries/get-a-backend.md): Returns a single backend by id.
- [List registries](https://docs.neuraltrust.ai/trustgate/api-reference/registries/list-registries.md): Returns a paginated list of registries in a gateway.
- [Test a backend connection](https://docs.neuraltrust.ai/trustgate/api-reference/registries/test-a-backend-connection.md): Validates connectivity and credentials against the provider's API with a lightweight, auth-only request. Test either a stored registry (registry_id) or an inline candidate configuration (provider + auth). Always returns 200; inspect "ok" and "stage" for the outcome.
- [Update a backend](https://docs.neuraltrust.ai/trustgate/api-reference/registries/update-a-backend.md): Updates an existing registry.
- [Attach a registry to a role](https://docs.neuraltrust.ai/trustgate/api-reference/roles/attach-a-registry-to-a-role.md): Associates a registry with a role (idempotent).
- [Create a role](https://docs.neuraltrust.ai/trustgate/api-reference/roles/create-a-role.md): Creates a new role in a gateway. model_policies cannot be set on create; bind registries first, then update the role.
- [Delete a role](https://docs.neuraltrust.ai/trustgate/api-reference/roles/delete-a-role.md): Deletes a role from a gateway.
- [Detach a registry from a role](https://docs.neuraltrust.ai/trustgate/api-reference/roles/detach-a-registry-from-a-role.md): Removes the association between a registry and a role (idempotent).
- [Get a role](https://docs.neuraltrust.ai/trustgate/api-reference/roles/get-a-role.md): Returns a role by id within a gateway.
- [List roles](https://docs.neuraltrust.ai/trustgate/api-reference/roles/list-roles.md): Returns a paginated list of roles in a gateway.
- [Update a role](https://docs.neuraltrust.ai/trustgate/api-reference/roles/update-a-role.md): Updates a role. model_policies may only reference registries already attached to the role.
- [Build version](https://docs.neuraltrust.ai/trustgate/api-reference/system/build-version.md): Returns build/version information for the running binary.
- [Liveness probe](https://docs.neuraltrust.ai/trustgate/api-reference/system/liveness-probe.md): Reports whether the process is alive.
- [Readiness probe](https://docs.neuraltrust.ai/trustgate/api-reference/system/readiness-probe.md): Reports whether the process is ready to serve traffic.
- [Admin API](https://docs.neuraltrust.ai/trustgate/api/overview.md): The TrustGate Admin API manages gateways, registries, consumers, auth, policies, roles, and catalogs. Authenticated with a bearer admin JWT.
- [Architecture](https://docs.neuraltrust.ai/trustgate/architecture.md): TrustGate is one Go binary that boots one of three independent planes — Admin, Proxy, MCP — backed by Postgres, Redis, and Kafka. How a request flows end to end.
- [Auth](https://docs.neuraltrust.ai/trustgate/concepts/auth.md): Auth credentials authenticate clients as a consumer — API key, OAuth2, OIDC, or mTLS. Created per gateway and attached to consumers.
- [Consumers](https://docs.neuraltrust.ai/trustgate/concepts/consumers.md): A consumer is the calling application's identity — it owns routing, credentials, and model policies, and its slug is the proxy URL path segment.
- [Gateways](https://docs.neuraltrust.ai/trustgate/concepts/gateways.md): A gateway is the top-level tenant in TrustGate — addressed by a slug, it owns the registries, consumers, auth, policies, and roles beneath it.
- [Registries](https://docs.neuraltrust.ai/trustgate/concepts/registries.md): A registry is an upstream backend — an LLM provider endpoint or an MCP server — with its own credentials, options, and health checks.
- [Roles](https://docs.neuraltrust.ai/trustgate/concepts/roles.md): Roles power identity-based routing: an OIDC token's claims select a role, and the role decides which registries, models, and MCP tools the caller may use.
- [Installation](https://docs.neuraltrust.ai/trustgate/getting-started/install.md): Install and run TrustGate with the one-line installer, Docker Compose, local dev, or Kubernetes.
- [Quickstart](https://docs.neuraltrust.ai/trustgate/getting-started/quickstart.md): From zero to a forwarded completion: create a gateway, register a provider, add a consumer and key, and call the OpenAI-compatible proxy.
- [MCP plane](https://docs.neuraltrust.ai/trustgate/mcp/overview.md): TrustGate's MCP plane aggregates registered Model Context Protocol servers into one endpoint — composing tools, prompts, and resources — with toolkit scoping, flexible upstream auth, and a built-in OAuth2 server for agents.
- [Metrics worker](https://docs.neuraltrust.ai/trustgate/observability/metrics.md): How TrustGate records a per-request metrics event and ships it off the hot path through an asynchronous worker — no scraping, no request-path overhead.
- [Telemetry](https://docs.neuraltrust.ai/trustgate/observability/telemetry.md): TrustGate streams a rich per-request event to Kafka — model, tokens, cost, latency breakdown, policy chain, and routing attempts — and can forward to TrustLens.
- [Configuration](https://docs.neuraltrust.ai/trustgate/operate/configuration.md): TrustGate is configured entirely from environment variables — servers, datastores, telemetry, timeouts, and discovery. The full reference.
- [Deployment](https://docs.neuraltrust.ai/trustgate/operate/deployment.md): Run TrustGate's planes as independently scalable processes via Docker or Kubernetes, backed by Postgres, Redis, and Kafka.
- [Server security](https://docs.neuraltrust.ai/trustgate/operate/server-security.md): How TrustGate secures its own surface: admin authentication, response security headers, and per-upstream client TLS.
- [Overview](https://docs.neuraltrust.ai/trustgate/overview.md): TrustGate is NeuralTrust's open-source, high-performance data-plane gateway for LLM and agent traffic — multi-provider routing, load balancing, policies, and MCP, in a single Go binary.
- [CORS](https://docs.neuraltrust.ai/trustgate/policies/cors.md): Allow browser clients to call the gateway with the cors policy, plus the server-level CORS middleware.
- [Cost cap](https://docs.neuraltrust.ai/trustgate/policies/cost-cap.md): A stateless per-request guard that rejects or downgrades any model whose list price per 1k tokens exceeds a configured ceiling.
- [Guardrails](https://docs.neuraltrust.ai/trustgate/policies/guardrails.md): Inspect prompts and responses inline at the gateway with TrustGuard, OpenAI Moderation, Azure Content Safety, or AWS Bedrock guardrail policies.
- [Policies](https://docs.neuraltrust.ai/trustgate/policies/overview.md): A policy is a configured governance rule on a gateway's traffic — rate limits, budgets, caching, tool governance, guardrails, and more — run at request/response stages, scoped globally or per consumer.
- [Prompt & model controls](https://docs.neuraltrust.ai/trustgate/policies/prompt-model-controls.md): Shape requests before they reach the model: inject and version system prompts with prompt_template, and restrict which models a consumer may call with model_allowlist.
- [Rate limiting & budgets](https://docs.neuraltrust.ai/trustgate/policies/rate-limiting.md): Limit request volume with rate_limiter and cap LLM spend with token_rate_limiter (LLM Budget) — token or dollar budgets, aggregate or per-model, scoped per consumer or gateway.
- [Request size](https://docs.neuraltrust.ai/trustgate/policies/request-size.md): Reject oversized requests before they reach a provider with the request_size_limiter policy — by payload size and character count.
- [Semantic cache](https://docs.neuraltrust.ai/trustgate/policies/semantic-cache.md): Cut cost and latency by serving cached responses for exact or semantically similar prompts, with a Redis, pgvector, or in-memory vector store.
- [Tool governance](https://docs.neuraltrust.ai/trustgate/policies/tool-governance.md): Govern agent tool use at the gateway: allow-list tools, reshape tool definitions, rate-limit individual tool calls, and validate the tool calls a model returns.
- [Fallback](https://docs.neuraltrust.ai/trustgate/routing/fallback.md): Retry across registries when an upstream fails — configurable triggers, an attempt/latency budget, and an ordered fallback chain.
- [Load balancing](https://docs.neuraltrust.ai/trustgate/routing/load-balancing.md): Distribute traffic across a consumer's registries with round-robin, weighted, least-connections, random, or semantic strategies — with health checks and weights.
- [Model resolution](https://docs.neuraltrust.ai/trustgate/routing/model-resolution.md): How the model field in a request selects a registry — short pass-through, provider-qualified, or pool reference — and how model policies constrain it.
- [Guard API](https://docs.neuraltrust.ai/trustguard/api/guard.md): POST /v1/guard — the single runtime endpoint. Request and response contract, authentication, attachments and SSRF, and status codes.
- [Collectors](https://docs.neuraltrust.ai/trustguard/concepts/collectors.md): A collector is a traffic tap — a gateway, SDK, browser extension, or WAF — that calls /v1/guard. It authenticates with an API key and routes each request to a policy.
- [Detectors](https://docs.neuraltrust.ai/trustguard/concepts/detectors.md): A detector is a reusable, configured instance of a catalog detector. It is detection-only — it decides what it finds, never what to do. Enforcement is decided by the policy that references it.
- [Policies](https://docs.neuraltrust.ai/trustguard/concepts/policies.md): A policy is where enforcement lives: gates that match request attributes before detection, detector rules that run detectors and decide the action, and a policy-wide Report/Enforce switch.
- [Agent & MCP security detectors](https://docs.neuraltrust.ai/trustguard/detectors/agent-mcp-security.md): Secure agentic and MCP traffic: scan the agent's own system prompt and tool/function descriptions for planted jailbreaks and prompt injections with tool_guard.
- [Application security detectors](https://docs.neuraltrust.ai/trustguard/detectors/application-security.md): Dangerous code/injection pattern detection for AI request payloads with code_sanitation.
- [Behavioral security detectors](https://docs.neuraltrust.ai/trustguard/detectors/behavioral-security.md): Detect abusive actors — bot-like cadence, repeated payloads, and malicious escalation — across requests and collectors with anomaly_detector.
- [Content security detectors](https://docs.neuraltrust.ai/trustguard/detectors/content-security.md): Jailbreak and prompt-injection detection, toxicity, topic moderation, and document/URL analysis on AI input and output.
- [Data loss prevention](https://docs.neuraltrust.ai/trustguard/detectors/data-loss-prevention.md): Detect and mask 43 PII entities and 4 secret types in AI traffic in flight, on input and output. The only transform-capable detector.
- [Detector catalog](https://docs.neuraltrust.ai/trustguard/detectors/overview.md): The built-in TrustGuard detector catalog: nine detectors across five security categories, with their supported sides and protocols.
- [How it works](https://docs.neuraltrust.ai/trustguard/how-it-works.md): The /v1/guard pipeline: resolve the collector from its API key, resolve the policy, run gates, run the detector rules, and reduce everything to a single status the caller enforces.
- [Application integrations](https://docs.neuraltrust.ai/trustguard/integrations/application.md): Call TrustGuard from your own code — the Python and Node SDKs, framework middleware, or the raw REST API — around your model calls.
- [Browser](https://docs.neuraltrust.ai/trustguard/integrations/browser.md): Inspect third-party AI web apps from a browser extension collector — Chrome, Edge, or Firefox.
- [Edge / WAF](https://docs.neuraltrust.ai/trustguard/integrations/edge-waf.md): Run TrustGuard enforcement at the CDN edge with Cloudflare Workers, AWS CloudFront (Lambda@Edge), Fastly Compute, or Akamai EdgeWorkers.
- [Gateway integrations](https://docs.neuraltrust.ai/trustguard/integrations/gateway.md): Run TrustGuard behind an AI gateway — TrustGate (first-class), Portkey, LiteLLM, Kong, Apigee, or Azure APIM — with the exact configuration each uses.
- [Integrations overview](https://docs.neuraltrust.ai/trustguard/integrations/overview.md): Every way to send traffic to TrustGuard — gateway, application SDK, browser, and edge/WAF collectors — all using the same /v1/guard call.
- [Overview](https://docs.neuraltrust.ai/trustguard/overview.md): TrustGuard is NeuralTrust's runtime security service: it evaluates AI traffic against the collector's policy and reports findings — without ever dropping traffic.
- [Alerts](https://docs.neuraltrust.ai/trustlens/alerts.md): Rule-based alerts that notify your team when an AI resource crosses a risk threshold — so posture regressions never sit unnoticed.
- [Data handling](https://docs.neuraltrust.ai/trustlens/data-handling.md): What TrustLens collects, what it never collects, where it stores data, and how to revoke access.
- [How it works](https://docs.neuraltrust.ai/trustlens/how-it-works.md): The five stages of the TrustLens lifecycle — connect, discover, assess, monitor, and alert — and how data flows from your environment into a single AI inventory.
- [Azure](https://docs.neuraltrust.ai/trustlens/integrations/azure.md): Azure AI Foundry (v2), AI Hub (classic Foundry), Azure OpenAI Classic (v1), and legacy ML Workspaces — all under one integration.
- [Endpoint Discovery (MDM)](https://docs.neuraltrust.ai/trustlens/integrations/endpoint-mdm.md): Deploy a read-only Device Discovery script via Microsoft Intune or Kandji to inventory AI tools — IDEs, browsers, extensions, agent CLIs, MCP servers, and agent configs — across macOS and Windows endpoints.
- [GCP Vertex AI](https://docs.neuraltrust.ai/trustlens/integrations/gcp-vertex-ai.md): Discover and monitor Vertex AI Reasoning Engines, models, datasets, and Model Armor guardrails across your Google Cloud project.
- [GitHub](https://docs.neuraltrust.ai/trustlens/integrations/github.md): Discover agent configs, MCP server definitions, and agent source code across your GitHub organization using a read-only GitHub App.
- [M365 Copilot & Copilot Studio](https://docs.neuraltrust.ai/trustlens/integrations/m365-copilot.md): Discover Copilot Studio bots via Dataverse and Microsoft 365 Copilot agents via the Microsoft Graph Agent Registry.
- [Mistral AI](https://docs.neuraltrust.ai/trustlens/integrations/mistral.md): Discover and monitor Mistral agents, models, files, document libraries, and native moderation guardrails using a single API key.
- [Inventory](https://docs.neuraltrust.ai/trustlens/inventory.md): The unified catalog of every AI surface across your organization — agents, models, datasets, IDEs, browsers, extensions, agent CLIs, MCP servers, agent configs, and managed endpoints.
- [Overview](https://docs.neuraltrust.ai/trustlens/overview.md): Discover and continuously assess every AI surface across your organization — agents, models, IDEs, browsers, MCP servers, and managed endpoints — from a single inventory.
- [Risk & findings](https://docs.neuraltrust.ai/trustlens/risk-and-findings.md): How TrustLens scores posture — the security controls evaluated per resource type, their weights, and how individual control results roll up to a 0–100 score and a risk level.
- [Trusttest sample code](https://docs.neuraltrust.ai/trusttest-sample-code.md)
- [Connect to NeuralTrust](https://docs.neuraltrust.ai/trusttest/connect/client.md)
- [Custom Targets](https://docs.neuraltrust.ai/trusttest/connect/custom.md)
- [HTTP Target](https://docs.neuraltrust.ai/trusttest/connect/http.md)
- [LLMs & Embeddings](https://docs.neuraltrust.ai/trusttest/connect/llms.md)
- [Overview](https://docs.neuraltrust.ai/trusttest/connect/overview.md)
- [Evaluation](https://docs.neuraltrust.ai/trusttest/core-concepts/evaluation-scenarios.md)
- [Overview](https://docs.neuraltrust.ai/trusttest/core-concepts/overview.md)
- [Test Generation](https://docs.neuraltrust.ai/trusttest/core-concepts/test-cases.md)
- [Agentic Behavior](https://docs.neuraltrust.ai/trusttest/create/agentic-behavior.md)
- [Test Generation](https://docs.neuraltrust.ai/trusttest/create/automatic-test-generation.md)
- [Content Bias](https://docs.neuraltrust.ai/trusttest/create/content-bias.md)
- [Creating Custom Probes](https://docs.neuraltrust.ai/trusttest/create/creating-custom-probes.md): Developer guide for building custom attack probes in TrustTest
- [Crescendo](https://docs.neuraltrust.ai/trusttest/create/crescendo.md)
- [From Dataset](https://docs.neuraltrust.ai/trusttest/create/dataset.md)
- [Echo Chamber](https://docs.neuraltrust.ai/trusttest/create/echo-chamber.md)
- [From Dataset](https://docs.neuraltrust.ai/trusttest/create/functional/from-dataset.md): Create functional tests from existing question-answer datasets
- [From Prompt](https://docs.neuraltrust.ai/trusttest/create/functional/from-prompt.md): Generate functional tests dynamically using LLM-powered prompt generation
- [From RAG](https://docs.neuraltrust.ai/trusttest/create/functional/from-rag.md): Generate functional tests from your Retrieval-Augmented Generation knowledge base
- [Overview](https://docs.neuraltrust.ai/trusttest/create/functional/overview.md): Evaluate your AI model's functional correctness and quality
- [Input Leakage](https://docs.neuraltrust.ai/trusttest/create/input-leakage.md)
- [Capture the Flag](https://docs.neuraltrust.ai/trusttest/create/iterate.md)
- [Azure](https://docs.neuraltrust.ai/trusttest/create/knowledge-base/connectors/azure.md)
- [In memory](https://docs.neuraltrust.ai/trusttest/create/knowledge-base/connectors/in-memory.md)
- [Neo4j](https://docs.neuraltrust.ai/trusttest/create/knowledge-base/connectors/neo4j.md)
- [Postgres](https://docs.neuraltrust.ai/trusttest/create/knowledge-base/connectors/postgres.md)
- [Upstash](https://docs.neuraltrust.ai/trusttest/create/knowledge-base/connectors/upstash.md)
- [Overview](https://docs.neuraltrust.ai/trusttest/create/knowledge-base/overview.md)
- [Off-Topics](https://docs.neuraltrust.ai/trusttest/create/off-topic.md)
- [Overview](https://docs.neuraltrust.ai/trusttest/create/overview.md)
- [From Prompt](https://docs.neuraltrust.ai/trusttest/create/prompt-dataset.md)
- [Prompt Injections](https://docs.neuraltrust.ai/trusttest/create/prompt-injections.md)
- [Sensitive Data Leak](https://docs.neuraltrust.ai/trusttest/create/sensitive-data-leak.md)
- [System Prompt Disclosure](https://docs.neuraltrust.ai/trusttest/create/system-prompt-disclosure.md)
- [Overview](https://docs.neuraltrust.ai/trusttest/create/threat-detection/overview.md): Complete reference of all TrustTest attack probes and techniques
- [From Dataset](https://docs.neuraltrust.ai/trusttest/create/threat-detection/prompt-injections/from-dataset.md): Load prompt injection attacks from curated or custom datasets
- [Crescendo Attack](https://docs.neuraltrust.ai/trusttest/create/threat-detection/prompt-injections/multi-turn/crescendo.md): Test resistance to gradual escalation attacks across multiple turns
- [Echo Chamber Attack](https://docs.neuraltrust.ai/trusttest/create/threat-detection/prompt-injections/multi-turn/echo-chamber.md): Test resistance to reinforcement-based manipulation through positive feedback
- [Multi-Turn Manipulation](https://docs.neuraltrust.ai/trusttest/create/threat-detection/prompt-injections/multi-turn/multi-turn-manipulation.md): Test resistance to gradual conditioning across multiple conversation turns
- [Multi-Turn Attacks](https://docs.neuraltrust.ai/trusttest/create/threat-detection/prompt-injections/multi-turn/overview.md): Sophisticated attacks across multiple conversation turns
- [Prompt Injections Overview](https://docs.neuraltrust.ai/trusttest/create/threat-detection/prompt-injections/overview.md): Complete catalog of prompt injection attack techniques in TrustTest
- [Best-of-N Jailbreaking](https://docs.neuraltrust.ai/trusttest/create/threat-detection/prompt-injections/single-turn/best-of-n.md): Test against multiple jailbreak variations to find model weaknesses
- [DAN Jailbreak](https://docs.neuraltrust.ai/trusttest/create/threat-detection/prompt-injections/single-turn/dan-jailbreak.md): Test resistance to "Do Anything Now" persona-based jailbreak attacks
- [Single Turn Attacks](https://docs.neuraltrust.ai/trusttest/create/threat-detection/prompt-injections/single-turn/overview.md): Direct prompt injection attacks delivered in a single message
- [Unsafe Outputs](https://docs.neuraltrust.ai/trusttest/create/unsafe-outputs.md)
- [Evaluation Context](https://docs.neuraltrust.ai/trusttest/evaluate-result/evaluation-strategy.md)
- [BLEU](https://docs.neuraltrust.ai/trusttest/evaluate-result/heuristics/bleu.md)
- [Equals](https://docs.neuraltrust.ai/trusttest/evaluate-result/heuristics/equals.md)
- [Language](https://docs.neuraltrust.ai/trusttest/evaluate-result/heuristics/language.md)
- [Overview](https://docs.neuraltrust.ai/trusttest/evaluate-result/heuristics/overview.md)
- [Regex](https://docs.neuraltrust.ai/trusttest/evaluate-result/heuristics/regex.md)
- [Completeness](https://docs.neuraltrust.ai/trusttest/evaluate-result/llm-as-a-judge/completeness.md)
- [Correctness](https://docs.neuraltrust.ai/trusttest/evaluate-result/llm-as-a-judge/correctness.md)
- [Custom](https://docs.neuraltrust.ai/trusttest/evaluate-result/llm-as-a-judge/custom.md)
- [Overview](https://docs.neuraltrust.ai/trusttest/evaluate-result/llm-as-a-judge/overview.md)
- [RAG Poisoning](https://docs.neuraltrust.ai/trusttest/evaluate-result/llm-as-a-judge/rag-poisoning.md)
- [Tone](https://docs.neuraltrust.ai/trusttest/evaluate-result/llm-as-a-judge/tone.md)
- [True/False](https://docs.neuraltrust.ai/trusttest/evaluate-result/llm-as-a-judge/true-false.md)
- [URL Correctness](https://docs.neuraltrust.ai/trusttest/evaluate-result/llm-as-a-judge/url-correctness.md)
- [Overview](https://docs.neuraltrust.ai/trusttest/evaluate-result/overview.md)
- [Installation](https://docs.neuraltrust.ai/trusttest/getting-started/installation.md)
- [Overview](https://docs.neuraltrust.ai/trusttest/getting-started/overview.md)
- [Quickstart](https://docs.neuraltrust.ai/trusttest/getting-started/quickstart.md)
- [Run Basic Red Teaming](https://docs.neuraltrust.ai/trusttest/getting-started/tutorials/basic-red-teaming.md)
- [Save and load Scenarios](https://docs.neuraltrust.ai/trusttest/getting-started/tutorials/client.md)
- [Run Responsibility Evaluation](https://docs.neuraltrust.ai/trusttest/getting-started/tutorials/compliance.md)
- [Custom LLM as a Judge](https://docs.neuraltrust.ai/trusttest/getting-started/tutorials/custom-llm-judge.md)
- [Http Target](https://docs.neuraltrust.ai/trusttest/getting-started/tutorials/http-model.md)
- [Run Capture The Flag Evaluation](https://docs.neuraltrust.ai/trusttest/getting-started/tutorials/iterate.md)
- [LLM as a Judge](https://docs.neuraltrust.ai/trusttest/getting-started/tutorials/llm-as-judge.md)
- [Quickstart with Local LLM](https://docs.neuraltrust.ai/trusttest/getting-started/tutorials/local-llm.md)
- [Test generation with Prompt](https://docs.neuraltrust.ai/trusttest/getting-started/tutorials/prompt-dataset.md)
- [Test generation with RAG](https://docs.neuraltrust.ai/trusttest/getting-started/tutorials/rag.md)

## OpenAPI Specs

- [openapi](https://docs.neuraltrust.ai/trustgate/api/openapi.json)

## Optional

- [Blog](https://neuraltrust.ai/blog)
- [Github](https://github.com/NeuralTrust)
- [Community](https://neuraltrustcommunity.slack.com/join/shared_invite/zt-2xl47cag6-_HFNpltIULnA3wh4R6AqBg)