Kestra Outputs and Metrics¶

Kestra is output-driven: flows hand off data through script outputs, files, and metrics. Truthound's Kestra package embraces that by exposing KestraOutputHandler, send_check_result, and related helpers rather than inventing a non-Kestra reporting surface.

Who This Is For¶

Kestra operators surfacing quality results in flow outputs
platform teams wiring metrics and summaries into downstream systems
engineers debugging what a script task actually emitted

When To Use It¶

Use this page when:

a Python script task should expose a structured quality result
a downstream task or dashboard needs the output
metrics and summaries must be emitted in a Kestra-friendly form

Prerequisites¶

truthound-orchestration[kestra] installed
a Kestra flow using Truthound script helpers
familiarity with the team's output and metrics conventions

Minimal Quickstart¶

Use the script entry point for the check:

from truthound_kestra import check_quality_script

result = check_quality_script(
    data_uri="s3://warehouse/curated/users.parquet",
    rules=[{"column": "id", "check": "not_null"}],
)

Then emit the result through the Kestra output helper:

from truthound_kestra import send_check_result

send_check_result(result)

Production Pattern¶

Use a layered output model:

Layer	Purpose
shared Truthound result	canonical status, counts, and details
Kestra output handler	task output transport and formatting
metrics and notifications	operational signals for the wider platform

Recommended practices:

emit one stable output contract per task
keep alerting and dashboard logic downstream from the shared result
distinguish raw detail from summarized metrics

Failure Modes and Troubleshooting¶

Symptom	Likely Cause	What To Do
downstream task cannot parse the output	result was formatted ad hoc	standardize on `send_check_result` or `KestraOutputHandler`
flow metrics exist but lack quality detail	only task success/failure is recorded	publish summarized counts from the result payload
different flows emit different result shapes	multiple custom handlers were introduced	converge on one shared output pattern