# Agentic security with Snyk Studio

Agentic workflows accelerate software development and innovation but introduce security risks, for example, AI-generated vulnerabilities or providing insecure practice suggestions. To address these risks, Snyk Studio embeds directives directly into AI-assisted development workflows to guide the agent towards secure software development.

Snyk Studio connects the Snyk platform, your development environment, and your AI tools, in four interconnected layers:

* [Snyk MCP Server](https://docs.snyk.io/discover-snyk/getting-started/glossary#snyk-mcp-server): An MCP server that enables easy integration with coding assistants, providing security context to AI agents. This runs locally using the Snyk CLI.
* AI agents or [ADE's](https://docs.snyk.io/discover-snyk/getting-started/glossary#ade): Agentic development environments that enable developers to interact with coding agents, which can integrate with Snyk Studio to drive security.
* [Directives](https://docs.snyk.io/discover-snyk/getting-started/glossary#directive): Policies, rules and commands provided to coding assistants to guide them on coding and security standards.
* The Snyk platform: Security intelligence from Snyk products (Snyk Code, Snyk Open Source, and more) which can be used to identify security issues and drive resolutions.

## MCP Server supported tools

Snyk Studio supports integrating the following Snyk security tools into an AI system:

* `snyk_sca_scan` (Open Source scan)
* `snyk_code_scan` (Code scan)
* `snyk_iac_scan` (IaC scan)
* `snyk_container_scan` (Container scan)
* `snyk_sbom_scan` (SBOM file scan)
* `snyk_aibom` (create AI-BOM)
* `snyk_trust` (trust a given folder before running a scan)
* `snyk_auth` (authentication)
* `snyk_logout` (logout)
* `snyk_version` (version information)
* `snyk_send_feedback` (summarizing issues fixed)
* `snyk_package_health_check` (evaluate package health during dependency selection)

{% hint style="info" %}
Running `snyk_sca_scan` can execute third-party ecosystem tools (for example, Gradle or Maven) on your machine to fetch the project's dependency tree.
{% endhint %}

## Coding assistant support

Snyk supports any AI agent and ADE that integrates with a local MCP server and offers [quickstart guides](https://docs.snyk.io/integrations/snyk-studio-agentic-integrations/quickstart-guides-for-snyk-studio) for some of these to provide additional guidance.

## Resources

Snyk provides a general setup guide with instructions applicable to all coding assistants. To learn more, visit [Getting started with Snyk Studio](https://docs.snyk.io/integrations/snyk-studio-agentic-integrations/getting-started-with-snyk-studio).

To learn about Snyk directives which govern agent behaviour and are automatically injected into agent interactions, visit [Guardrail Directives](https://docs.snyk.io/integrations/directives#guardrail-directives). To learn about Snyk directives which can be manually invoked by human developers or AI agents, visit [Command Directives](https://docs.snyk.io/integrations/directives#command-directives).

Snyk Studio supports large-scale distribution and centralized governance. For guidance on how to handle scaling across your organization, visit [Distribution at scale](https://docs.snyk.io/integrations/snyk-studio-agentic-integrations/distribution-at-scale).


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.snyk.io/integrations/snyk-studio-agentic-integrations.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
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.