Use this file to discover all available pages before exploring further.
MCP resources are read-only data endpoints that give AI clients a structured snapshot of your workspace. The Navis Ops MCP server exposes four static resources and one templated resource. All resources require mcp:read scope (or mcp:write).Resources are fetched on demand — every time your AI client reads a resource, it gets the current data at that moment. There are no live subscriptions or push updates. If you need fresh data after making changes, your client must re-read the resource.
To list available resources, use resources/list. For templated resources, use resources/templates/list to get the URI template, then construct the URI with your values.
These four resources are always available. Their URIs are fixed — no parameters required.
navis://workspace/summary
Workspace overviewA compact summary of your entire workspace. Returns aggregate counts and recent activity so an AI assistant can quickly orient itself before deciding which tools to call.Returns:
Total project count
Total task count (all statuses)
Total note count
Total calendar event count
Total goal count
Recent activity (last modified items across types)
Use this when you want the assistant to get a high-level picture of your workspace before doing more specific work.
Templated resources use URI templates (RFC 6570). Fill in the parameter to construct a valid URI, then read it with resources/read.
navis://projects/{id}/tasks
Open tasks for a specific projectURI template: navis://projects/{id}/tasksReturns the open tasks for a single project, identified by its ID. Filters to non-completed tasks and returns up to 100 results.Parameters:
Parameter
Type
Description
id
string (UUID)
The project ID. Use list_projects or the navis://projects/all resource to look up project IDs.
Returns:
Project metadata (id, name, status)
Up to 100 open tasks for that project: id, title, status, priority, due_date, assignee
Example URI:navis://projects/3fa85f64-5717-4562-b3fc-2c963f66afa6/tasksUse this when the assistant is working within the context of a specific project and needs its task list.To get the URI template via JSON-RPC:
Resources are fetched on demand — they reflect the current state of your workspace at the time of the read. There are no live subscriptions or change notifications. If your AI assistant needs up-to-date data after making changes (for example, after calling create_task), it must re-read the relevant resource.
If you’re using an AI assistant interactively and notice it’s working from stale data, ask it to re-read the relevant resource. For example: “Re-read navis://tasks/today and tell me what’s overdue.”