Cortex MCP server

What the model can do

Cortex developer portal: catalog entities (services, resources, domains, teams) with owner teams and owners, members, Slack channels, groups and repository, one entity's links, hierarchy and Scorecard scores, teams with their members, Scorecards with each entity's score, level and failing rules, deploys in a time window, and an entity's custom events. When asked, it can add a custom event, a deploy record or a custom data value to an entity, record a dependency, or archive or unarchive an entity.

Reads

• cortex_entitiesCortex catalog entities, a page at a time: tag, name, type, description, owner teams and owner emails, members, Slack channels, groups…
• cortex_searchRelevance-ranked search of the Cortex catalog (words with AND, OR, NOT): tag, name, type, owner team tags and owner emails…
• cortex_entityOne Cortex entity in depth: description, owner teams, groups, repository, links, parent and child entities, and its score…
• cortex_teamsCortex teams: tag, name, summary, type (Cortex-defined or from the identity provider), members by name and email, Slack channels, whether archived…
• cortex_scorecardsCortex Scorecards: tag, name, description, whether a draft, its level names, how many rules, and when created and last updated.
• cortex_scorecard_scoresScores on one Cortex Scorecard: for each entity, its score, percentage, level, the rules it fails and when it was last evaluated.
• cortex_deploysDeploys, rollbacks, scales and restarts recorded in Cortex in a time window: title, type, environment, SHA, who deployed (name and email)…
• cortex_custom_eventsCustom events on one Cortex entity's timeline (incidents, releases, anything an integration posted): title, type, link and time.

Acts, when you ask

These change something in Cortex. The model is told to use them only when you ask for exactly that, and each one is granted by name: per team on a shared connection, by the owner on a personal one.

• cortex_add_custom_eventAdd a custom event (an incident, a release note, a migration milestone) to a Cortex entity's timeline.
• cortex_add_deployRecord a deploy, rollback, scale or restart against a Cortex entity.
• cortex_set_custom_dataSet a custom data value on a Cortex entity under a key (Scorecard rules can read it).
• cortex_add_dependencyRecord in Cortex that one entity depends on (calls) another.
• cortex_archive_entityArchive a Cortex entity: it leaves the catalog lists and Scorecards until it is unarchived.
• cortex_unarchive_entityUnarchive a Cortex entity so it is back in the catalog and Scorecards.

Connect it

First connect Cortex on Dashboard → Connections, then add Memtro to your client once. The Cortex tools appear alongside the tools of everything else you have connected; nothing is configured per connector.

claude plugin marketplace add memtro-org/claude-memtro
claude plugin install memtro@memtro --config api_key=sk-memtro-...

{
  "mcpServers": {
    "memtro": {
      "url": "https://platform.memtro.com/api/mcp",
      "headers": {
        "Authorization": "Bearer sk-memtro-..."
      }
    }
  }
}

{
  "mcp": {
    "memtro": {
      "type": "remote",
      "url": "https://platform.memtro.com/api/mcp",
      "headers": {
        "Authorization": "Bearer sk-memtro-..."
      }
    }
  }
}

Who may use it

The key identifies the person, and the server offers that person the tools they may use everywhere else in Memtro. A shared Cortex connection is used by the teams an administrator granted it to, tool by tool, on the Permissions page; a personal connection is used by its owner alone, and its tools that act stay off until the owner allows them. See Teams and permissions and Connectors.

Ask it like this

• › Which services are failing our production readiness Scorecard?
• › Who owns the checkout service?
• › What was deployed to production yesterday?
• › Which rules is the payments service failing?
• › Archive the old reporting service

Questions

Is this an official Cortex MCP server?

No. It is Memtro's MCP server with Memtro's Cortex connector behind it: the connector reads Cortex through its API with the credentials you connect on Dashboard → Connections, and Cortex publishes nothing here. The same endpoint carries every other system you connect and Memtro's memory tools.

What does it cost?

Nothing on top of Memtro. One price per organisation by its number of users, from $50 a month for up to 10 users after a 7-day free trial, with every connector, the MCP server, jobs and memory included. You bring your own model keys.

Can it change anything in Cortex?

Only when you ask, and only what it has been granted: add a custom event to an entity's timeline; record a deploy against an entity (records it, does not deploy); set a custom data value on an entity (may replace one with the same key); record that one entity depends on another; archive an entity (hidden from the catalog and Scorecards until unarchived); unarchive an entity. Every other tool reads, searches or lists. An administrator grants each acting tool by name per team, and a personal connection acts only after its owner allows it.

Which clients work?

Any MCP client that speaks Streamable HTTP: Claude Code, Cursor, Claude Desktop, OpenCode, Codex, Zed, Windsurf, Cline and the Anthropic and OpenAI SDKs. There is no local process to run; the server is remote and authenticated with your key.

Other MCP servers · Engineering and operations

• GitHub MCP server
• AWS CloudWatch MCP server
• Sentry MCP server
• ServiceNow MCP server
• Okta MCP server
• Twilio MCP server