Cortex MCP Server
Connect Cortex to your AI agents with the Cortex MCP server.
Updated
What is Cortex MCP?
The Cortex MCP server is a Model Context Protocol server that connects Cortex to AI agents and assistants as a set of tools. With it connected, an agent can work with Cortex (internal developer portal) directly — instead of you copying data in and out by hand.
It is a remote MCP server that uses a bearer token, so you point your client at the server URL, authorize once, and the Cortex tools become available to the model.
Connect the Cortex MCP server
Connect Cortex MCP to Claude Code
- 1
Add the server
Use `claude mcp add` with the Cortex MCP server URL (or its `npx` command), following the official docs.
- 2
Authorize
Complete authentication (a bearer token) so Claude Code can call the server.
- 3
Verify
Ask Claude to use a Cortex tool to confirm the connection is live.
Connect Cortex MCP to Cursor
- 1
Open MCP settings
In Cursor, go to Settings → MCP → Add new server.
- 2
Add Cortex
Provide the Cortex MCP server URL or command and complete authentication.
- 3
Test
Reference Cortex from chat to confirm Cursor can reach the tools.
Sample use cases
An agent queries Cortex for service ownership and maturity scores, then generates a team health report
Automated visibility into service ownership gaps and engineering standards compliance.
A scheduled agent checks Cortex scorecards for services that dropped below threshold and alerts the owning team
Proactive engineering quality enforcement without manual scorecard reviews.
An agent looks up the on-call owner and runbook for a service during an incident
Faster incident response by surfacing the right contact and documentation immediately.
An agent audits service catalog entries for missing documentation, ownership, or dependencies
Cleaner service catalog with fewer orphaned or undocumented services.
An agent correlates Cortex service data with deploy frequency and incident rates for an engineering metrics dashboard
DORA-style metrics enriched with service ownership and maturity context.
An agent creates a new service entry in Cortex when a repository is initialized, pre-populating ownership and metadata
Every new service starts with correct catalog metadata from day one.
An agent pulls dependency graphs from Cortex to identify blast radius before a breaking infrastructure change
Better change management by understanding downstream impact before deploying.
Using Cortex MCP with Gamut
On Gamut, the Cortex MCP server becomes a tool your agents use inside an automated, event-driven workflow — triggered by an event or a schedule, acting through Cortex and your other connected MCP servers, with the connection managed by Gamut rather than a local config file.
Frequently asked questions
What is Cortex and what does its MCP server provide?
Cortex is an internal developer portal that tracks service ownership, scorecards, and engineering standards. The MCP server gives agents access to the service catalog, maturity scores, team ownership, dependencies, and runbooks.
Can the Cortex MCP server help with incident response?
Yes. Agents can look up service ownership, on-call schedules, runbooks, and dependency maps during an incident, reducing the time to identify the responsible team and relevant documentation.
Does the Cortex MCP server support scorecard automation?
Yes. Agents can read scorecard results and maturity scores for any service. This enables workflows like alerting teams when scores drop, generating compliance reports, or tracking improvement over time.
Can I use Cortex MCP alongside monitoring tools?
Yes. In Gamut, agents can combine Cortex (for service ownership and metadata) with monitoring MCP servers like Datadog or PagerDuty to build enriched incident workflows that know both what broke and who owns it.