Name: gamussa/flink-sql
Rating: 0.958 (1 reviews)
Author: gamussa

Blog Docs Log in Get started

gamussa/flink-sql

Apache Flink SQL, Table API, and UDF development for both OSS Flink and Confluent Cloud

1.21x

Review — 95%

Does it follow best practices?

Evaluation — 97%

↑ 1.21x

Agent success when using this tile

Validation — 11 / 11 Passed

Validation for skill structure

Evaluation results

100%

Real-Time Sensor Monitoring Dashboard

Window aggregations and watermarks

Criteria

Without context

With context

Tumbling window TVF

100%

Hopping window TVF

100%

Session window TVF

100%

Cumulate window TVF

100%

Sensor watermark declaration

100%

Activity watermark declaration

100%

Energy watermark declaration

100%

Window GROUP BY includes window_start/end

100%

Kafka connector config

100%

Aggregation functions

100%

DESCRIPTOR references correct column

100%

Partition by sensor/user/meter

100%

Without context: $0.4193 · 1m 37s · 12 turns · 61 in / 8,241 out tokens

With context: $0.5358 · 1m 33s · 15 turns · 63 in / 6,659 out tokens

100%

16%

GDPR Data Masking UDFs for Confluent Cloud

UDF development and deployment

Criteria

Without context

With context

Extends ScalarFunction

100%

Public eval method

100%

Null handling in scalar

100%

Extends TableFunction

100%

FunctionHint annotation

100%

Collect method usage

100%

Maven shade plugin

100%

Flink dependency provided scope

100%

Python udf decorator

100%

Artifact upload command

100%

CREATE FUNCTION with artifact

100%

LATERAL TABLE usage

100%

Without context: $0.3538 · 1m 22s · 14 turns · 14 in / 5,739 out tokens

With context: $0.9337 · 2m 41s · 31 turns · 78 in / 9,977 out tokens

95%

E-Commerce Order Enrichment Pipeline

Join patterns and CDC handling

Criteria

Without context

With context

Interval join syntax

100%

Temporal join syntax

100%

Lookup join with PROCTIME

100%

Debezium CDC format

100%

CDC primary key

100%

Upsert-kafka connector

100%

Upsert primary key

100%

Deduplication pattern

100%

Watermarks on streams

100%

Currency rates primary key

100%

State TTL consideration

No unbounded regular join

100%

Without context: $0.5455 · 2m 48s · 10 turns · 59 in / 13,238 out tokens

With context: $0.9246 · 2m 59s · 29 turns · 74 in / 12,363 out tokens

100%

43%

Real-Time Fraud Detection with Stateful Processing

Process Table Functions

Criteria

Without context

With context

Extends ProcessTableFunction

100%

SET_SEMANTIC_TABLE annotation

100%

StateHint with TTL

20%

100%

ValueState for per-key state

100%

OnTimer annotation

40%

100%

Timer registration

87%

100%

FunctionHint output type

100%

SQL PARTITION BY

100%

SQL uid parameter

100%

SQL on_time DESCRIPTOR

37%

100%

CAST for time parameter

42%

100%

Collect Row output

100%

Without context: $0.9613 · 4m 39s · 20 turns · 20 in / 19,901 out tokens

With context: $1.9494 · 6m 25s · 39 turns · 4,683 in / 26,160 out tokens

94%

24%

Confluent Cloud Flink Pipeline with Table API

Confluent Cloud deployment

Criteria

Without context

With context

ConfluentSettings usage

100%

ConfluentTools for results

25%

Statement Set syntax

80%

100%

MATCH_RECOGNIZE structure

100%

System column $rowtime

100%

Compute pool create command

100%

Statement create command

100%

Savepoint workflow

100%

Exception list command

100%

Tumbling window for metrics

100%

Confluent Cloud plugin dependency

25%

100%

No unsupported features

42%

100%

Without context: $0.7170 · 2m 56s · 23 turns · 71 in / 13,045 out tokens

With context: $0.7681 · 2m 16s · 28 turns · 327 in / 8,064 out tokens

Install with Tessl CLI

npx tessl i gamussa/flink-sql@1.0.0

Evaluated: 3 days ago
Agent: Claude Code

Table of Contents

Real-Time Sensor Monitoring Dashboard GDPR Data Masking UDFs for Confluent Cloud E-Commerce Order Enrichment Pipeline Real-Time Fraud Detection with Stateful Processing Confluent Cloud Flink Pipeline with Table API