> For the complete documentation index, see [llms.txt](https://docs.solcis.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.solcis.ai/where-solcis-enforces.md). # Where Solcis enforces Where your controls are enforced: the Solcis Policy Engine on the action path, gateways, and NVIDIA OpenShell and Microsoft AGT, all governed from Solcis Compliance Studio. Solcis enforces your adopted controls on what your AI systems do. Solcis Compliance Studio sends enforcement policies to enforcement points and collects their decisions and evidence back. The Solcis Policy Engine returns Allow, Deny or Escalate, each with a severity rating, before the action happens. The Solcis Policy Engine writes a compliance receipt for each decision; Compliance Studio's universal evidence ledger collects receipts from every enforcement point. Every enforcement point applies the same Compliance Studio policies, so the choice is only about *where* enforcement sits, not what it enforces or how it's recorded. ## Solcis Policy Engine The Solcis Policy Engine enforces on the action path, and at the model, tool and MCP endpoints your agents call. With the Solcis SDK, the Policy Engine is built directly into your AI applications and agents, at the point where they decide to act, for more complete coverage and greater accuracy. Because it runs inside the application it sees the full picture: which agent, which user, what it is trying to do and why. It also governs actions that stay inside the app and never cross the network. Enforcement is built in as the AI is built, so compliance ships with the product rather than being bolted on later. ## Gateways Your own gateway or the Solcis Gateway is an inline checkpoint in the path of your AI's traffic, in front of the models, tools and MCP servers your agents call. When an application calls a model, or an agent calls a tool, an MCP server, or an API, that call passes through the gateway and is evaluated against your controls, authored in Compliance Studio as enforcement policy, before it is allowed to proceed. There is no change to application code. New applications route through the same gateway as you build them, so coverage grows without integration work and your engineers stay focused on shipping product. It is a fast way to bring AI under enforcement. ## NVIDIA OpenShell and Microsoft AGT NVIDIA OpenShell and Microsoft AGT enforce security and infrastructure controls in the agent runtime. Compliance Studio can supply them enforcement policies and collect their decisions and evidence. ## One policy set, many enforcement points A gateway is the fast first line; the Solcis Policy Engine, built in with the Solcis SDK, is the deeper layer you build in as your AI matures. Because policies come from Compliance Studio, moving between enforcement points is not a rebuild: your controls, policies and evidence ledger stay in Compliance Studio. An action that reaches no enforcement point isn't evaluated or recorded; Compliance Studio's coverage and agent discovery show where the gaps are. | | Solcis Policy Engine (with Solcis SDK) | Gateway (Solcis or your own) | NVIDIA OpenShell and Microsoft AGT | | -------------------------- | -------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------- | | What it is | The Solcis Policy Engine built into your AI applications and agents | An inline checkpoint your AI's traffic passes through | Agent runtimes that enforce security and infrastructure controls | | Where it sits | Inside your application or agent, on the action path | In front of the models, tools and MCP servers your agents call | In the agent runtime | | Change to your application | Built into the application by your team | Route traffic through it; no application changes | Compliance Studio can supply enforcement policies to the runtime | | Coverage | The action path and the model, tool and MCP endpoints it's placed on, including actions that never cross the network | Calls routed through it | Agents running in that runtime | | Best for | More complete coverage and accuracy | Bringing AI under enforcement fast, across many applications | Security and infrastructure controls in the agent runtime, with policies supplied and evidence collected by Compliance Studio | Shared by every enforcement point: policies from Compliance Studio and one universal evidence ledger in Compliance Studio. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://docs.solcis.ai/where-solcis-enforces.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.