Connect the Snyk MCP server to Unblocked to give AI agents live AI security context

Teams adopting AI ship models, agents, MCP servers, and skills faster than anyone can track them, and Evo by Snyk is where that inventory, the policies governing it, and the violations against those policies live. Without a connection, an agent adding a new model or MCP server has no idea whether it breaks a policy until someone checks Evo by hand. Snyk’s Evo MCP server exposes that data to AI agents, and Unblocked connects to it once so every agent can ask without its own Snyk setup.

This is a third-party MCP server: Snyk hosts and maintains it, and Unblocked connects to it as a verified MCP server. Nothing is ingested or indexed — Unblocked calls the read-only Evo tools you select at the moment a question needs them, so answers reflect your AI inventory and open issues as Evo reports them now.

Wiring the Evo MCP server straight into an agent gives it raw asset and issue queries. Unblocked’s Context Engine reasons across those results and your other sources — tracing a policy violation on an MCP server to the repository that uses it, the pull request that introduced it, and the Slack thread where someone asked for an exception — and returns one cited answer.

Once connected, any MCP-compatible coding agent — Claude Code, Cursor, Devin, GitHub Copilot, Codex — reaches Snyk context through a single Unblocked MCP server connection.

What you get

• Live AI inventory and policy issues — Agents query the models, agents, MCP servers, skills, and packages Evo has discovered, along with the policies and violations attached to them, straight from Snyk’s hosted server.
• MCP allow list checks before you add a server — Agents can look up an MCP server by name, package, or URL and read your organization’s allow list, so the question of whether a server is approved gets answered in the editor.
• Read-only, tool-level control — Evo’s policy and allow list write tools are never offered. You choose which read tools Unblocked may call, and Snyk still enforces your Tenant role on every request. Change the selection anytime from Settings → Data Sources.
• Context Engine reasoning — Evo results are reconciled with your code, pull requests, tickets, and conversations, so a flagged AI asset leads to the change that introduced it and the people who discussed it — with citations.

Prerequisites

• An Unblocked account (free to start).
• A Snyk account with access to Evo by Snyk, a Tenant role that can read Evo data, and permission to authorize Snyk’s OAuth sign-in.
• An MCP-compatible coding agent (Claude Code, Cursor, Devin, etc.) to use the context in your editor.

Step 1: Connect Snyk to Unblocked

1. In Unblocked, go to Settings → Data Sources → Connect another data source, and under Security & Compliance select Snyk.
2. The Snyk MCP server URL is filled in for you. Choose OAuth, click Connect, and sign in to Snyk to authorize Unblocked.
3. Once Unblocked discovers the server’s tools, select the read-only tools it may use and click Save Settings.

Step 2: Install the Unblocked MCP server

On macOS or Linux, one command auto-detects and configures the supported agents on your machine:

On Windows, or when you prefer remote MCP, add the server manually using the configuration below, or follow the manual install guide .

claude mcp add -s user --transport http unblocked https://getunblocked.com/api/mcpsse

{
  "mcpServers": {
    "unblocked": {
      "url": "https://getunblocked.com/api/mcpsse"
    }
  }
}

https://getunblocked.com/api/mcpsse

Snyk hosts and maintains this MCP server. Unblocked calls only the read-only tools you select, at the moment a question needs them:

• query_assetsReturns AI assets and their metadata — models, agents, MCP servers, skills, prompts, datasets, packages, and more — from your Evo inventory.
• query_related_assetsReturns relationships between assets, including indirect ones, such as which repositories or machines use a given model or MCP server.
• query_policiesReturns policy definitions and metadata, including each policy’s conditions and whether it is a Snyk default or a custom policy.
• query_issuesReturns policy violations, with the violating asset and the location where it was found.
• find_approvable_mcp_serversLooks up MCP servers by name, package name, or URL, covering servers detected in your environment and known servers in Snyk’s catalogue.
• get_mcp_server_allow_listReturns your organization’s current MCP server allow list.

To give Claude Code, Cursor, GitHub Copilot, Codex, or another MCP-compatible AI coding agent access to Snyk context, connect Snyk to Unblocked and add the single Unblocked MCP server to your agent.

Unblocked acts as the MCP context layer for your Snyk data, searching AI assets, policies, and policy violations alongside your other connected sources and returning cited results through these tools:

• context_researchResearches a question across your connected sources and calls the Snyk Evo tools you selected when AI assets, policies, or policy violations are relevant, returning a synthesized answer with citations.
• context_get_urlsFetches the full content behind URLs your agent encounters across connected sources — Jira, Linear, Notion, Confluence, Slack, GitHub, and more.

Questions you (or your agent) can ask Unblocked about your Snyk data:

• Is the MCP server this pull request adds on our allow list, and has anyone asked for it in Slack?
• Which open Evo issues involve models used in the payments repository, and who introduced them?
• What does our policy on unapproved model providers say, and which recent changes would break it?
• Which agents and skills did Evo find on developer machines, and is there a ticket tracking their review?