Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Prefect Flow Builder · dicklesworthstone | DecimalAI

dicklesworthstone/prefect-flow-builder

SkillScore100Maintained

Build prefect flow builder operations. Auto-activating skill for Data Pipelines. Triggers on: prefect flow builder, prefect flow builder Part of the Data Pipelines skill category. Use when working with prefect flow builder functionality. Trigger with phrases like "prefect flow builder", "prefect builder", "prefect".

by dicklesworthstone

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

Data Engineer pack

1 · One-time — install the CLI (docs ↗)

pip install decimalai

2 · Get the skill — writes .claude/skills/dicklesworthstone-prefect-flow-builder/SKILL.md

decimalai skills pull dicklesworthstone-prefect-flow-builder --out .claude/skills/

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/dicklesworthstone-prefect-flow-builder/SKILL.md https://app.decimal.ai/s/dicklesworthstone-prefect-flow-builder/SKILL.md

Raw file: /s/dicklesworthstone-prefect-flow-builder/SKILL.md ↗ · machine-readable summary at /s/dicklesworthstone-prefect-flow-builder.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 23d ago

Yes — +18% lift+19% tokens0% turns100% of whole cases pass with skill (22/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-07	✗→✓	▲ Improved	6%	0%
case-09	✗→✓	▲ Improved	31%	0%
case-12	✗→✓	▲ Improved	19%	0%
case-13	✗→✓	▲ Improved	21%	0%
case-01	✓→✓	＝ Same ✓	60%	0%

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/10/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

dicklesworthstone / pi_agent_rust

tests/ext_conformance/artifacts/plugins-community/skills/11-data-pipelines/prefect-flow-builder/SKILL.md

View on GitHub ↗

License MITImported 8/10/2026Version v1

Skill contents

SKILL.md

Prefect Flow Builder

Overview

This skill provides automated assistance for prefect flow builder tasks within the Data Pipelines domain.

When to Use

This skill activates automatically when you:

Mention "prefect flow builder" in your request
Ask about prefect flow builder patterns or best practices
Need help with data pipeline skills covering etl, data transformation, workflow orchestration, and streaming data processing.

Instructions

Provides step-by-step guidance for prefect flow builder
Follows industry best practices and patterns
Generates production-ready code and configurations
Validates outputs against common standards

Examples

Example: Basic Usage Request: "Help me with prefect flow builder" Result: Provides step-by-step guidance and generates appropriate configurations

Prerequisites

Relevant development environment configured
Access to necessary tools and services
Basic understanding of data pipelines concepts

Output

Generated configurations and code
Best practice recommendations
Validation results

Error Handling

| Error | Cause | Solution | |-------|-------|----------| | Configuration invalid | Missing required fields | Check documentation for required parameters | | Tool not found | Dependency not installed | Install required tools per prerequisites | | Permission denied | Insufficient access | Verify credentials and permissions |

Resources

Official documentation for related tools
Best practices guides
Community examples and tutorials

Related Skills

Part of the Data Pipelines skill category. Tags: etl, airflow, spark, streaming, data-engineering

Illustrative Want to see what dicklesworthstone/prefect-flow-builder would put in front of your model? Preview the rendered body and check its trigger phrases — no LLM call, no signup.

Latest benchmark · publisher runVerified

8/30/2026, 04:38 UTC · 22 cases

Lift

+18%

100% vs 82%

Agent turns

1.0 vs 1.0

Tokens

+19%

2,362.95 vs 1,985.73

fail → pass · 4pass → pass · 18

Per-case results (22)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-01 Write a Python script defining an entrypoint function for an orchestration workflow that logs a starting message. You might be tempted to use plain Python functions or Airflow DAG context managers, but construct standard Prefect flow decorators.	pass→pass	8,993	10,220	+14%	1	1	0%	728	1,162	+60%	0	0	—
case-02 Define a data extraction step and a data transformation step inside a Python pipeline script. Base models often put all execution logic directly inside a single flow function, but separate distinct unit operations using task decorators.	pass→pass	16,381	18,669	+14%	1	1	0%	2,498	3,251	+30%	0	0	—
case-03 Create an API fetching task that must retry up to 3 times with a 5-second delay upon failure. Models often recommend writing manual try-except retry loops in standard Python; use native decorator keyword arguments instead.	pass→pass	15,850	11,936	-25%	1	1	0%	2,200	1,708	-22%	0	0	—
case-04 Configure a dynamic run name for a data ingestion pipeline function that includes the parameter 'dataset_name' and execution date. Instead of setting global environment variables, pass a dynamic string template to the flow definition.	pass→pass	18,903	17,102	-10%	1	1	0%	2,481	2,654	+7%	0	0	—
case-05 Define input parameters for a flow that processes batch records, enforcing typed validation for an integer batch size and string endpoint URL. Do not use manual kwargs checking; leverage standard type annotations or Pydantic models in the flow parameters.	pass→pass	9,118	15,473	+70%	1	1	0%	1,839	2,374	+29%	0	0	—
case-06 Write an asynchronous data scraping task that queries a HTTP endpoint using httpx. Async tasks in workflow frameworks often default to blocking calls; ensure proper async syntax is used with the task decorator.	pass→pass	13,670	18,445	+35%	1	1	0%	2,572	2,983	+16%	0	0	—
case-07 Design a master orchestrator function that triggers a child workflow function for processing sub-components. Avoid calling subprocesses directly; call the secondary flow directly inside the primary flow execution context.	fail→pass	22,207	19,063	-14%	1	1	0%	4,014	4,254	+6%	0	0	—
case-08 Configure a flow that processes multiple independent S3 files in parallel using concurrent task execution. Base models often default to sequential execution; explicitly set a concurrent task runner on the flow.	pass→pass	17,030	18,219	+7%	1	1	0%	2,384	3,128	+31%	0	0	—
case-09 Configure a database query task so its result is cached for 1 hour based on the input parameter key. Rather than using custom Redis caching code, use the framework's native cache keys and expiration settings.	fail→pass	10,293	13,041	+27%	1	1	0%	2,270	2,964	+31%	0	0	—
case-10 Attach a alert notification handler function to fire whenever a pipeline execution enters a Failed state. Avoid wrapping every line in try/catch; use hook arguments directly on the decorator.	pass→pass	15,836	7,464	-53%	1	1	0%	2,029	1,871	-8%	0	0	—
case-11 Create a deployment specification for a flow named 'etl-process' targeted at a work pool named 'prod-pool'. Construct a valid deployment yaml definition matching standard orchestration manifests.	pass→pass	9,251	10,421	+13%	1	1	0%	1,443	2,274	+58%	0	0	—
case-12 Generate a summary report table at the end of a pipeline run and publish it to the UI run dashboard. Do not print raw stdout; create a markdown artifact using the official artifact API.	fail→pass	16,490	19,689	+19%	1	1	0%	2,407	2,869	+19%	0	0	—
case-13 Retrieve a runtime configuration value named 'DATABASE_TIMEOUT' inside a task. Instead of using `os.getenv`, read it from the centralized orchestration variable store.	fail→pass	13,151	6,403	-51%	1	1	0%	1,217	1,471	+21%	0	0	—
case-14 Fetch an API key securely stored in a Secret block named 'stripe-api-key' inside a task. Avoid hardcoding plaintext or using standard config files; load the block instance programmatically.	pass→pass	12,723	8,909	-30%	1	1	0%	1,551	1,837	+18%	0	0	—
case-15 Implement a manual approval step in a deployment flow before executing a database schema migration. Base models often use `input()` which hangs non-interactive runners; use the framework's pause mechanism.	pass→pass	19,756	25,992	+32%	1	1	0%	2,628	4,461	+70%	0	0	—
case-16 Ensure a long-running data transformation task aborts if it exceeds 300 seconds. Avoid custom signal handling; use native decorator arguments for timeout limits.	pass→pass	17,355	11,556	-33%	1	1	0%	2,036	1,720	-16%	0	0	—
case-17 Iterate over a list of items and submit a task for each item concurrently, gathering the futures afterwards. Do not call task functions directly in a sync loop; use the `.submit()` method.	pass→pass	11,455	6,969	-39%	1	1	0%	1,199	1,740	+45%	0	0	—
case-18 Limit concurrent executions of a rate-limited API task across all runs to a maximum of 5 concurrent slots using concurrency limit blocks or contexts. Do not use python threading semaphores.	pass→pass	17,089	12,696	-26%	1	1	0%	2,267	2,836	+25%	0	0	—
case-19 Log an INFO level message inside a task function. Base models usually use standard `print()` or `logging.getLogger()`; obtain the dedicated runtime logger context.	pass→pass	14,082	8,546	-39%	1	1	0%	1,586	1,025	-35%	0	0	—
case-20 Author an Apache Airflow DAG script that schedules a daily PySpark job using the `BashOperator` and `DAG` context manager. Do not use Prefect constructs.	pass→pass	11,881	6,938	-42%	1	1	0%	1,409	1,741	+24%	0	0	—
case-21 Write a pure PySpark script to read a parquet file from S3, perform a filter operation on 'age > 21', and write back to S3. Do not include workflow orchestrator decorators or flows.	pass→pass	5,629	12,084	+115%	1	1	0%	1,216	1,724	+42%	0	0	—
case-22 Write a dbt SQL model file that performs an incremental aggregation of daily sales transactions from a raw schema. Do not write Python or orchestration flow code.	pass→pass	13,506	13,414	-1%	1	1	0%	1,712	1,938	+13%	0	0	—

These cases were also screened by the skill’s intent / safety pass.See the Trust review ↗Reproduce this benchmark in Colab ↗

How this skill was measured

DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +18 percentage points is the difference between those two pass rates over the 22 comparable cases.

Model measured: gemini-3.6-flash
Graded by: gemini-3.6-flash
Grading method: judged
Verification: verified
Pass rate without the skill: 81.8%
Pass rate with the skill: 100%
Cases attempted: 22
Cases counted toward lift: 22
Cases that regressed: 0
Skill version measured: v1
Run completed: 8/30/2026, 04:38 UTC

Measured example — same prompt, with vs. without the skill · from the benchmark run · case case-07

Prompt

Design a master orchestrator function that triggers a child workflow function for processing sub-components. Avoid calling subprocesses directly; call the secondary flow directly inside the primary flow execution context.

Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.

Show both answers

SkillScore blends 4 signals1 of 4 complete

BenchmarkPass rate WITH skill on a verified run100

Live evalNeeds ≥10 evaluated traces in 30 days

AI ratingNeeds ≥10 rated traces

AdoptionNeeds ≥10 serves across ≥2 orgs in 30 days

Efficiency · separate axis (at equal correctness)+0% turns+20% tokens

Shown for transparency; not credited unless a verified run shows no correctness regression.

Backed by 1 of 4 signals — each added signal makes the score more robust.

Verified on gemini-3.6-flash · re-tested 22 days ago

Other measured skills in the registry, with their headline benchmark lift.