---

name: runner-and-programmatic description: "Guides programmatic Metaflow execution with Runner, NBRunner, Deployer bridge objects, subprocess status, logs, async waits, and returned run/task inspection." disable-model-invocation: true metadata: disco-role: operating license: Apache 2.0

Runner and Programmatic Execution

Use this sub-skill when a task asks to run a Metaflow flow from Python code, automate a local smoke test, inspect ExecutingRun/ExecutingTask, stream logs, or bridge to deployment objects programmatically.

Quick Route

• Read references/runner-api.md for verified Runner usage, kwargs placement, returned objects, status, logs, and artifact access.
• Read references/notebook-and-spin.md for notebook and Spin-adjacent APIs.
• Read references/troubleshooting.md when subprocesses fail, time out, or return an unexpected Run lookup error.
• Run scripts/runner_smoke.py for a self-contained local programmatic smoke test.

Minimal Programmatic Pattern

from metaflow import Runner

with Runner("flow.py", show_output=False, pylint=False, env={"USERNAME": "disco"}) as runner:
    executing = runner.run(max_workers=1)
    print(executing.status, executing.returncode)
    print(executing.run.pathspec)

Top-level flow-script options such as pylint=False, profile=..., env=..., or datastore=... belong on Runner(...). Run-command options such as max_workers=1, tags=[...], or run_id_file=... belong on runner.run(...).

Boundaries

• For writing the flow file itself, route to ../flow-authoring/SKILL.md.
• For detailed object traversal after a run exists, route to ../client-and-data/SKILL.md.
• For real Argo or Step Functions deployment prerequisites, route to ../deployment-orchestration/SKILL.md.