Tool reference · Tasks, workflows & decisions
workflow_learn
Stores a named, ordered list of steps as a reusable workflow so it can be predicted and tracked later.
local stdio server writes
When to use
- You just finished a multi-step procedure (a release, a migration) that will be repeated.
- You want a stable workflow_id to pass to workflow_track after each run.
- An existing workflow with the same name and project needs new steps.
Parameters
| Name | Type | Default | Description |
|---|---|---|---|
namerequired | string | — | — |
stepsrequired | string[] | — | — |
description | string | — | — |
trigger_pattern | string | — | — |
context | object | — | — |
project | string | "general" | — |
Example
Arguments
{
"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",
"trigger_pattern": "release",
"project": "my-library"
}Result shape
{
"workflow_id": "3f9c2a7e5b1d4c8f9a0e6b2d7c4f1a8e"
}Saving again with the same name and project updates that workflow and returns the same id.
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:
Record a learned workflow (named sequence of steps) for future reuse.