OpsLevel MCP server
OpsLevel in any MCP client, through Memtro: OpsLevel does not publish this server. Memtro's OpsLevel connector reads OpsLevel through its API with the credentials you connect, and one endpoint hands those tools, every other system you connect and Memtro's memory tools to Claude Code, Cursor, Claude Desktop or anything else that speaks Streamable HTTP, as you.
What the model can do
OpsLevel service catalog: services with owner team, tier, lifecycle, maturity level, language and tags; teams with members and contacts; one service's note, failing and passing checks by category, dependencies, dependents and last deploy; teams; deploys in a time window; scorecards with pass rates; repositories. When asked, it can replace a service's note, add or remove tags, create, edit or delete a service, and record a dependency between services.
Reads
opslevel_servicesOpsLevel services in the catalog: name, aliases, type, owner team, tier, lifecycle, maturity level, product, language, framework…opslevel_serviceOne OpsLevel service in depth: owner, tier, lifecycle, tags (with ids), its note, maturity level by category, every check's latest result…opslevel_teamsOpsLevel teams: name, alias (what owner filters take), parent team, responsibilities, members by name, email and role (first 50)…opslevel_deploysDeploys OpsLevel recorded in a time window: service alias, environment, status, when it started and finished, how long it took…opslevel_scorecardsOpsLevel scorecards: name, owner team, how many services it applies to, checks passing out of total and the pass rate.opslevel_repositoriesRepositories OpsLevel knows: name, alias, owner team, tier, languages, default branch, whether archived or hidden, and links.opslevel_levelsThe tiers and lifecycles defined in this OpsLevel account, with the aliases that filters and service edits take.
Acts, when you ask
These change something in OpsLevel. 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.
opslevel_set_service_noteReplace an OpsLevel service's note (markdown shown on the service page) with new text.opslevel_add_tagsAdd key:value tags to an OpsLevel service.opslevel_remove_tagDelete a tag from an OpsLevel service, by the tag's id (from opslevel_service).opslevel_create_serviceCreate a service in the OpsLevel catalog, optionally with an owner team, tier, lifecycle, product, language and framework.opslevel_update_serviceChange an OpsLevel service's name, description, owner team, tier, lifecycle, product, language or framework.opslevel_add_dependencyRecord in OpsLevel that one service depends on another, with an optional note.opslevel_delete_serviceDelete a service from the OpsLevel catalog, with its tags, note, dependencies and check history.
Connect it
First connect OpsLevel on Dashboard → Connections, then add Memtro to your client once. The OpsLevel tools appear alongside the tools of everything else you have connected; nothing is configured per connector.
Claude Code
The Memtro plugin registers the server and a skill for when to use it; the tools are then available in every session.
claude plugin marketplace add memtro-org/claude-memtro
claude plugin install memtro@memtro --config api_key=sk-memtro-...Cursor
Add the server to .cursor/mcp.json in the project, or to the global file under ~/.cursor/.
{
"mcpServers": {
"memtro": {
"url": "https://platform.memtro.com/api/mcp",
"headers": {
"Authorization": "Bearer sk-memtro-..."
}
}
}
}Claude Desktop
Create a personal API key on Dashboard → Profile → API keys. In Claude Desktop add an MCP server of type HTTP with URL https://platform.memtro.com/api/mcp and the header Authorization: Bearer sk-memtro-…
OpenCode
A remote server in opencode.json.
{
"mcp": {
"memtro": {
"type": "remote",
"url": "https://platform.memtro.com/api/mcp",
"headers": {
"Authorization": "Bearer sk-memtro-..."
}
}
}
}Pi
One command: pi mcp add memtro --url https://platform.memtro.com/api/mcp --bearer-token-env-var MEMTRO_API_KEY. Tools arrive as mcp__memtro__*. The pi-memtro extension registers it for you, alongside Memtro as the model.
Any other client
Add an MCP server of type HTTP (Streamable HTTP, stateless) at https://platform.memtro.com/api/mcp with the header Authorization: Bearer sk-memtro-.... Codex, Zed, Windsurf, Cline and the SDKs all take a remote server this way.
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 OpsLevel 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 tier 1 services are failing checks?
- › Who owns the payments service, and how do I reach them?
- › What was deployed to production in the last 24 hours?
- › Which scorecards have the lowest pass rate?
- › Record that checkout depends on payments
Questions
- Is this an official OpsLevel MCP server?
- No. It is Memtro's MCP server with Memtro's OpsLevel connector behind it: the connector reads OpsLevel through its API with the credentials you connect on Dashboard → Connections, and OpsLevel 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 OpsLevel?
- Only when you ask, and only what it has been granted: replace a service's note (the previous note is returned); add tags to a service; remove a tag from a service; create a service in the catalog; change a service's name, description, owner team, tier, lifecycle, product, language or framework; record that one service depends on another; delete a service from the catalog. 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.