oncall-router-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| PYTHONPATH | Yes | The absolute path to the src directory of the oncall-router-mcp project, required to import the oncall_router.server module. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| who_ownsA | Which team owns a service and how to reach them right now. Accepts the name someone would actually type during an incident, including aliases. Returns found=false with suggestions rather than guessing at a near match. |
| escalation_pathA | Who to wake for this service at this severity, in order, with the minute each hop is due measured from impact start. Refuses an unknown severity rather than defaulting to the quietest one. |
| playbookA | What the runbook says to check first. Falls back to the service's general steps when the symptom is unknown, and says so with fell_back=true. |
| impact_clockB | Given when impact started, which escalation hop should be active now and which are overdue. Time is measured from impact start rather than ticket creation. Requires an explicit now timestamp: it never assumes the current time. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 4 tools
Each tool targets a distinct aspect of incident routing: ownership (who_owns), escalation (escalation_path), response steps (playbook), and timing (impact_clock). There is no overlap in purpose, and the descriptions make boundaries clear.
All tool names are lowercase snake_case and are descriptive, but the pattern varies slightly: who_owns is a question-like phrase, while the others are noun compounds. This is a minor inconsistency, but the naming remains intuitive and predictable.
With only 4 tools, the server is tightly scoped to incident routing. Each tool delivers a unique, essential capability without redundancy, making the count ideal for the domain.
The tool set covers the core incident routing workflow: identifying ownership, following escalation paths, accessing playbooks, and tracking escalation timing. Minor gaps like incident acknowledgment or status updates exist but are outside the router's stated purpose.