---
title: "Redpanda streaming: review, setup & limitations · Undominated.ai"
canonical: https://undominated.ai/skills/redpanda-streaming/
description: "Redpanda reference skill for using the broker through its Kafka-compatible API: rpk topic and record commands, producer and consumer settings, transactions, client compatibility, tiered storage and licensed enterprise topic features."
---

# Redpanda streaming: review, setup & limitations · Undominated.ai

> Redpanda reference skill for using the broker through its Kafka-compatible API: rpk topic and record commands, producer and consumer settings, transactions, client compatibility, tiered storage and licensed enterprise topic features.

[← Explore all skills](/skills/)

DATA PIPELINES / Redpanda

# Redpanda streaming

Redpanda reference skill for using the broker through its Kafka-compatible API: rpk topic and record commands, producer and consumer settings, transactions, client compatibility, tiered storage and licensed enterprise topic features.

 [See setup guidance ↓](#setup)[Original source ↗](https://github.com/redpanda-data/skills)

SOURCE REVIEW

 Reviewed 2026-10-21
 Evidence 5 linked sources
 Publisher Redpanda
 Licence [Apache-2.0 ↗](https://github.com/redpanda-data/skills/blob/a4562333b774d963066e11dfa1282646b8490550/LICENSE)
 Revision a4562333b774
 [Read what was—and wasn’t—checked ↓](#review)

“Teaches how to use the Redpanda broker through its Kafka-compatible API: producing and consuming data, consumer groups, idempotent producers, exactly-once semantics and transactions, topic creation and configuration, Kafka client compatibility, and tiered storage. …”

 [Redpanda · upstream description ↗](https://github.com/redpanda-data/skills/blob/a4562333b774d963066e11dfa1282646b8490550/skills/streaming/SKILL.md) Our analysis follows below.

01 / THE REASONING

## Why this made the selection.

 - SKILL.md works through the Kafka API with rpk commands (topic list, create, produce, consume, alter-config, delete) and a franz-go Go snippet, plus producer and consumer property tables, the transaction call sequence with isolation.level=read_committed, and tiered storage settings.
- references/SOURCES.md maps each file of the skill to the redpanda broker source paths and docs pages it derives from, and the README says the skills are kept in sync by scheduled generator, critic and human-merge routines with a monthly drift audit.
- It states compatibility limits: Kafka clients 0.11 or later, a list of validated clients, and that Redpanda does not implement KIP-890, so Kafka 4.x clients fall back to the original transaction protocol.

### A good fit for

 - Connect rpk to a local or SASL/SCRAM-secured Redpanda cluster and list, create, produce to and consume from topics
- Set producer and consumer properties such as acks, enable.idempotence, compression.type and group.id for a Kafka client against Redpanda
- Write a transactional producer and read only committed data with isolation.level=read_committed
- Enable tiered storage, Iceberg topics, cloud topics or shadow links on a cluster with an Enterprise licence

### Weigh up before choosing

 - skills/streaming last changed in a commit dated 2026-10-20 (an update to references/SOURCES.md); the pinned revision is dated 2026-10-20.
- The skill covers the Kafka API surface (port 9092 by default). The Admin HTTP API and the rpk CLI reference are in the separate streaming-admin-api and rpk skills.
- Tiered storage and the enterprise features SKILL.md describes (Iceberg topics, cloud topics, continuous data balancing, shadow linking) require a Redpanda Enterprise licence.
- references/SOURCES.md says the skill is grounded in the public redpanda-data/redpanda repository and in auto-generated reference and feature pages in the private redpanda-data/docs repository, read via the Redpanda-Github-Read MCP connector or gh, and that all cluster, topic and broker property claims are grounded in the auto-generated property partials. It defers Kafka API handler version ranges, release-specific metrics and third-party client config keys to live introspection rather than pinning them.
- Run as written, SKILL.md's commands change cluster state: rpk topic create, produce, alter-config and delete, rpk cluster config set, and rpk shadow create and failover.
- The SASL/SCRAM example passes the user name and password to rpk as -X user= and -X pass= flags on the command line.
- LICENSE is the Apache License 2.0 with the appendix placeholder "Copyright [yyyy] [name of copyright owner]" left unfilled; the README says the skills are distributed under the Apache 2.0 license.

02 / THE REVIEW RECORD

## What we actually inspected.

Source review has boundaries. A clear record is more useful than a “safe” badge.

### Material inspected

 - skills/streaming/SKILL.md at a4562333b774d963066e11dfa1282646b8490550
- skills/streaming/references/SOURCES.md at a4562333b774d963066e11dfa1282646b8490550
- README.md at a4562333b774d963066e11dfa1282646b8490550
- .claude-plugin/marketplace.json and .claude-plugin/plugin.json at a4562333b774d963066e11dfa1282646b8490550
- LICENSE at a4562333b774d963066e11dfa1282646b8490550

### Our findings

 - The skill directory holds SKILL.md and 15 reference files and no scripts; the SKILL.md frontmatter declares only name and description, with no allowed-tools.
- references/SOURCES.md maps every skill file to source paths in the public redpanda-data/redpanda repository and to docs pages in the private redpanda-data/docs repository, and tells maintainers to verify against the current stable release tag.
- The README prints no command that installs this skill alone. Both install routes add the redpanda plugin, whose marketplace source is the repository root: 32 SKILL.md files at this revision.
- Commands appear only as examples in fenced blocks, including topic deletion, cluster configuration changes and shadow-link failover; the skill bundles nothing to execute.

### Not established by this review

 - Not executed: it was not installed or run.
- No Redpanda cluster was contacted and no rpk command was run.
- No topic was created and no record was produced or consumed.

The review applies to the material and revision named here. A newer upstream release can change its behavior.

03 / PUT IT TO WORK

## Add a skill to your workflow.

[Upstream setup instructions ↗](https://github.com/redpanda-data/skills/blob/a4562333b774d963066e11dfa1282646b8490550/README.md)
 - The README's one-command line is `claude plugin marketplace add redpanda-data/skills && claude plugin install redpanda@redpanda-skills`. It installs the redpanda plugin, whose marketplace source is the repository root: all 32 SKILL.md files at this revision, not this skill alone.
- Inside a Claude Code session the README gives the same as two slash commands, run one at a time: `/plugin marketplace add redpanda-data/skills` and `/plugin install redpanda@redpanda-skills`.
- Once installed, the README says the skill is invoked as `/redpanda:streaming`, or loaded automatically when its frontmatter description matches the task.

### Before you start

 - Claude Code, for the README's plugin commands
- A reachable Redpanda cluster and rpk for any live command
- A Redpanda Enterprise licence for tiered storage and the enterprise topic features

### Compatibility

Claude Code · Kafka clients 0.11 or later

### Implementation

Go · Java

### Skill text; rpk and client commands run against the user's cluster

 - SKILL.md declares no allowed-tools and bundles no scripts; it is reference text with rpk command and client code examples.
- Its rpk examples read and write the user's cluster over the Kafka API: list, create, alter and delete topics; produce records; consume them, optionally as a consumer group.
- Cluster-level examples change cluster configuration (rpk cluster config set iceberg_enabled, partition_autobalancing_mode, enable_shadow_linking) and run rpk shadow create and failover.
- SASL/SCRAM credentials are passed to rpk as -X user= and -X pass= flags.

### Cost model

Apache-2.0 skill text. SKILL.md says tiered storage and the enterprise topic features require a Redpanda Enterprise licence; the files read state no price for that licence or for Redpanda Cloud.

04 / FOLLOW THE EVIDENCE

## The source trail.

Our notes are separate from the original resource. Check upstream before adopting a new version.

 - [streaming SKILL.md at a4562333 ↗](https://github.com/redpanda-data/skills/blob/a4562333b774d963066e11dfa1282646b8490550/skills/streaming/SKILL.md) Checked 2026-10-21 https://github.com/redpanda-data/skills/blob/a4562333b774d963066e11dfa1282646b8490550/skills/streaming/SKILL.md Supports: upstreamDescription, whySelected, bestFor, compatibility, language, limitations, access, review.findings, install
- [streaming references/SOURCES.md at a4562333 ↗](https://github.com/redpanda-data/skills/blob/a4562333b774d963066e11dfa1282646b8490550/skills/streaming/references/SOURCES.md) Checked 2026-10-21 https://github.com/redpanda-data/skills/blob/a4562333b774d963066e11dfa1282646b8490550/skills/streaming/references/SOURCES.md Supports: whySelected, limitations, review.findings
- [README.md at a4562333 ↗](https://github.com/redpanda-data/skills/blob/a4562333b774d963066e11dfa1282646b8490550/README.md) Checked 2026-10-21 https://github.com/redpanda-data/skills/blob/a4562333b774d963066e11dfa1282646b8490550/README.md Supports: install, compatibility, whySelected, limitations, review.findings
- [.claude-plugin/marketplace.json at a4562333 ↗](https://github.com/redpanda-data/skills/blob/a4562333b774d963066e11dfa1282646b8490550/.claude-plugin/marketplace.json) Checked 2026-10-21 https://github.com/redpanda-data/skills/blob/a4562333b774d963066e11dfa1282646b8490550/.claude-plugin/marketplace.json Supports: homepage, install, review.findings
- [Apache-2.0 LICENSE at a4562333 ↗](https://github.com/redpanda-data/skills/blob/a4562333b774d963066e11dfa1282646b8490550/LICENSE) Checked 2026-10-21 https://github.com/redpanda-data/skills/blob/a4562333b774d963066e11dfa1282646b8490550/LICENSE Supports: license, limitations

KEEP COMPARING

## Other approaches to consider.

Related by category or shared topics. These are alternatives to inspect, not a measured quality order.

 [### Developing Kafka Python client ↗ Confluent skill that scaffolds a confluent-kafka-python producer, consumer or both, or adds Schema Registry to existing Python code, for Confluent Cloud, local Docker or WarpStream, after a confirmation step its instructions require.](/skills/confluent-developing-kafka-python-client/)[### Elasticsearch File Ingest ↗ Load CSV or JSON into Elasticsearch with explicit mapping and bulk-result checks.](/skills/elastic-elasticsearch-ingest/)[### Elasticsearch Reindex Planning ↗ Prepare destination mappings and verify asynchronous local or remote reindex operations.](/skills/elastic-elasticsearch-reindex/)

 [AI Tools ↗](/tools/)[Skills ↗](/skills/)[Agents ↗](/agents/)[MCP Servers ↗](/mcp-servers/)[Workflows ↗](/workflows/)

## Continue your investigation

 - [Choose a toolkit](/tools/)
- [Inspect execution hosts](/agents/)
- [Check tool access](/mcp-servers/)
- [Check project dependencies](/check-repo/)
