get_run_output
Retrieve raw data for one explicitly requested named output or a bounded retained stdout/stderr page from a completed model run. Provide exactly one of output_name or stream. Do not use this for a request to see, show, view, or display a compatible tabular model output or financial statement; use create_output_sheet_view as the default Native Sheet presentation. Provides view: run-output (ui://apps/run-output) renders this tool’s result when MCP Apps are available.
read-only · idempotent
Parameters
Section titled “Parameters”| Name | Type | Required | Default | Description |
|---|---|---|---|---|
run_id | Annotated[str, Field(description='UUID of a completed model run, or a short UUID prefix. Use list_runs to discover run IDs.')] | Yes | — | |
output_name | `Annotated[str | None, Field(description=“Key in the run’s outputs dict to retrieve (e.g. ‘forecast.json’ or ‘revenue’). Provide exactly one of output_name or stream.”)]` | No | None |
encoding | Annotated[Literal['auto', 'json', 'text', 'base64'], Field(description="How to encode the returned content: 'auto' (default) sniffs content type; 'json' returns raw JSON; 'text' returns text; 'base64' returns base64-encoded bytes.")] | No | 'auto' | |
stream | `Annotated[Literal[‘stdout’, ‘stderr’] | None, Field(description=‘Retained run stream to retrieve in bounded pages. Provide exactly one of stream or output_name.’)]` | No | None |
projection | `Annotated[str | None, Field(description=‘RFC 6901 JSON Pointer evaluated within the selected named output. Use an empty string for its root. Valid only with output_name.’)]` | No | None |
offset | Annotated[int, Field(description='Zero-based retained-byte offset for a stream page, or item offset for a projected named-output page. Follow the prior continuation.')] | No | 0 | |
limit_items | Annotated[int, Field(description=f'Maximum array items or object members in a projected named-output page (1–{MAX_JSON_PROJECTION_ITEMS}). Ignored for streams and unprojected output.')] | No | DEFAULT_JSON_PROJECTION_ITEMS | |
limit_bytes | Annotated[int, Field(description=f'Maximum raw stream bytes or compact-JSON projected bytes in one page (4–{MAX_STREAM_PAGE_BYTES}). Ignored for unprojected named output.')] | No | DEFAULT_STREAM_PAGE_BYTES | |
workspace_id | `Annotated[str | None, Field(description=ID_BACKED_WORKSPACE_ID_FIELD_DESCRIPTION)]` | No | None |
Returns
Section titled “Returns”Returns a structured object. Top-level fields:
run_id(string)model_name(string)output_name(string)resolved_output_name(string)content(array<object>)encoding(string)size_bytes(integer)truncated(boolean)resolution_path(string)
See the response example below for the full payload shape.
Example
Section titled “Example”Tool Call
Section titled “Tool Call”{ "name": "get_run_output", "arguments": { "run_id": "66666666-6666-6666-6666-666666666666", "output_name": "revenue", "encoding": "auto" }}Response
Section titled “Response”{ "run_id": "66666666-6666-6666-6666-666666666666", "model_name": "forecasts", "output_name": "revenue", "resolved_output_name": "revenue", "content": [ { "month": "2026-01", "revenue": 100 }, { "month": "2026-02", "revenue": 110 } ], "encoding": "json", "size_bytes": 78, "truncated": false, "resolution_path": "literal"}- Module:
Execution - Requires an authenticated Bridge Town session (OAuth) or an API token.
- When
get_runreturnstruncated: true, call this tool directly with a name fromoutput.outputs_index— do not re-callget_runor ask the user which output to fetch first.