skills.rest
ricardolui
ricardoluiCOMMUNITY
Explore all skills by this author →
GitHub

confluent-skill-creator

Create and validate Confluent product skills tested against real Kafka environments.

Updated Jul 7, 2026
Search Tags:#e2e testing#kafka#cdc#confluent#flink#schema registry#skill creator
One-click install
npx skills add https://github.com/ricardolui/gcp-custom-agent-skills --skill confluent-skill-creator-ricardolui
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: confluent-skill-creator
Source: https://github.com/ricardolui/gcp-custom-agent-skills/tree/main/confluent-skill-creator
Command: npx skills add https://github.com/ricardolui/gcp-custom-agent-skills --skill confluent-skill-creator-ricardolui

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires confluent-kafka, requests, jsonschema, python-dotenv, and includes scripts (resource) and references (resource) components.

What problem does it solve? Building high-quality agent skills for Confluent products (Kafka, Flink, Schema Registry, Tableflow, Connectors) requires deep platform knowledge, spec compliance, and real end-to-end testing — this Skill guides the entire authoring workflow so generated skills are correct, secure, and validated against live Confluent environments. ## Core Features & Use Cases - Guided skill authoring workflow: Walks through scope definition, platform targeting (Confluent Cloud, Platform, Apache Kafka, WarpStream), interaction method selection (MCP, CLI, REST APIs, client libraries), and spec compliance validation. - Real E2E testing with bundled scripts: Ships Python scripts for producing JSON_SR data, consuming and verifying messages, registering schemas, checking Flink compute pools, and cleaning up resources. - Security guardrails built in: Enforces credential-file protection (never read .env contents), synthetic-data-only rules, and plan-before-execute confirmation for all resource-modifying operations. - Use Case: A developer wants to publish a "CDC from PostgreSQL to Iceberg via Debezium and Tableflow" skill. This Skill interviews them, generates the SKILL.md with evals and credential templates, validates it against the Agent Skills spec, and runs E2E tests against Confluent Cloud before packaging. ## Quick Start Ask the assistant to create a new Confluent skill, for example: "Create a skill for real-time order enrichment using Flink SQL on Confluent Cloud."

Frequently Asked Questions about confluent-skill-creator

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create a Confluent skill for Kafka or Flink?▼

Describe the use case or component scope, then answer the interview questions about platform targeting and interaction method (MCP, CLI, REST APIs, or client libraries). The workflow generates SKILL.md, evals, and credential templates, validates spec compliance, and runs E2E tests before packaging.

What platforms can a generated Confluent skill target?▼

Generated skills can target Confluent Cloud, Confluent Platform, Apache Kafka, or WarpStream. Platform-specific skills include the platform name in the skill name, while cross-platform skills ship per-platform reference files and a platform-detection step.

Does the skill creator test against real Confluent environments?▼

Yes, E2E tests run against real environments — local Docker, Confluent Platform, Confluent Cloud, or WarpStream — never mocks. Bundled scripts produce JSON_SR data, consume and verify messages, register schemas, and check Flink compute pool availability.

How are credentials handled when testing Confluent skills?▼

Credentials live in a .env or credentials.yaml file that is never read, displayed, or logged; scripts reference variable names only and verify existence with test -n. Both template formats are provided, and the credential file must be added to .gitignore.

What validation does a generated skill go through before delivery?▼

Generated skills are validated against the Agent Skills specification: frontmatter rules, description quality, trigger-overlap checks, SKILL.md under 500 lines, evals/evals.json with assertions, and a final skills-ref validate run before packaging.

When should I not use the Confluent skill creator?▼

Do not use it to directly operate Confluent products, such as building a pipeline, writing a producer, or deploying Flink SQL. It is only for authoring new skills; use the appropriate product-specific skill for direct Confluent work.

Related Skills

Explore More Software Engineering Skills→
confluentincconfluentinc
42

developing-kafka-python-client

Create a Python Kafka client project with Schema Registry and JSON Schema validation.

Official
Intermediate
dvy1987dvy1987
3

publish-skill

Package and publish validated agent skills to skills.sh and GitHub.

Community
Advanced
amadeus-dlcamadeus-dlc
5

skill-forge

Create and refine Claude Code and Codex skill packages with validation and iteration workflows.

Official
Advanced
armoin2018armoin2018
4

ailey-data-kafka

Automate Kafka workflows for real-time data pipelines across Confluent Cloud, Platform, Apache Kafka, and AWS MSK.

Community
Advanced