Tool reference · Tasks, workflows & decisions
workflow_predict
Estimates how likely a stored workflow is to succeed and how long it usually takes, based on its past runs.
local stdio server read-only
When to use
- Before starting a task that matches a workflow you have run before.
- You only remember a keyword such as "release" and want the best matching workflow.
- You want to check whether enough runs exist for the estimate to be trusted (see confidence).
Parameters
| Name | Type | Default | Description |
|---|---|---|---|
workflow_id | string | — | — |
trigger | string | — | — |
project | string | — | — |
Example
Arguments
{
"trigger": "release",
"project": "my-library"
}Result shape
{
"found": true,
"workflow_id": "3f9c2a7e5b1d4c8f9a0e6b2d7c4f1a8e",
"success_probability": 0.8333,
"raw_success_rate": 0.9,
"avg_duration_ms": 420000,
"times_run": 10,
"confidence": 0.6667,
"workflow": {
"name": "release-python-package",
"steps": [
"Run the full test suite",
"Bump version in pyproject.toml",
"Build wheel and sdist",
"Upload to PyPI"
],
"description": "Standard release for the core library"
}
}The probability is smoothed as (successes + 1) / (runs + 2). With no match, found is false and the other values are null.
Values are illustrative; the keys follow the server's handler. MCP clients receive the result as JSON text content.
Server description
The description the server sends to your agent in tools/list, captured from the v14.7.0 source:
Predict outcome (success probability, avg duration) for a workflow by id OR by trigger keyword. Uses Laplace-smoothed success rate.