A single-call tool that gathers everything you need before modifying code at a specific location. It bundles the function source, all callers that would be affected, related tests, relevant memories, and recent git history into one response.
Use stellarion_get_ai_context instead when you only need to understand or explain code without making changes.
| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
| uri | string | Yes | -- | File URI of the code being edited (e.g., file:///path/to/file.ts) |
| line | number | Yes | -- | Line number being edited (0-indexed) |
| maxTokens | number | No | 8000 | Maximum tokens of context to return. Priority order: symbol > callers > tests > memories > git history. |
Natural Language:
"I need to add a timeout parameter to fetchUserProfile"
MCP Tool Call:
{
"name": "stellarion_get_edit_context",
"arguments": {
"uri": "file:///home/dev/project/src/api/users.ts",
"line": 34
}
}
Returns: Full source of fetchUserProfile, all 7 callers that would need updating, 3 test functions that call it, a memory noting "fetchUserProfile was recently refactored to use async/await", and 2 recent commits touching this file.
Natural Language: "There's an off-by-one error in the pagination logic"
MCP Tool Call:
{
"name": "stellarion_get_edit_context",
"arguments": {
"uri": "file:///home/dev/project/src/db/pagination.rs",
"line": 72,
"maxTokens": 12000
}
}
Returns: Full source of the pagination function, callers from API handlers, existing test cases for boundary conditions, a known_issue memory about "pagination breaks on empty result sets", and git log showing the last 3 changes to the file.
The response contains 5 sections, filled within the token budget in priority order:
get_ai_context (4000) because it includes more context sections. Increase it further for complex functions.stellarion_get_symbol_info instead -- it returns just the metadata.