MCP Resources
In addition to tools, the MCP server provides resources — read-only context providers that give AI assistants structured data about your projects, sprints, and stories.
Resource URIs
| URI Pattern | Description |
|---|---|
sprintvibe://project/{id} |
Project overview with sprint summary and backlog count |
sprintvibe://sprint/{id} |
Sprint with all epics and stories |
sprintvibe://story/{id} |
Full story with description, acceptance criteria, and tasks |
How resources work
Resources are read via the resources/read JSON-RPC method. Your AI assistant typically handles this automatically when it needs context about a specific item.
Example request:
json
{
"jsonrpc": "2.0",
"id": 1,
"method": "resources/read",
"params": {
"uri": "sprintvibe://story/42"
}
}
Example response:
json
{
"jsonrpc": "2.0",
"id": 1,
"result": {
"contents": [
{
"uri": "sprintvibe://story/42",
"mimeType": "application/json",
"text": "{...full story JSON...}"
}
]
}
}
When to use resources vs. tools
- Resources are for getting context — your AI reads them to understand what you're working on
- Tools are for taking actions — listing, creating, updating, or commenting
In practice, most AI assistants use tools for everything. Resources provide an alternative way to fetch read-only context when the assistant supports the MCP resources protocol.