custom-kafka-consumer
Set up, configure, run, and debug a custom Kafka consumer that streams data into Snowflake via Snowpipe Streaming SDK v2. Use when: building a Kafka-to-Snowflake streaming pipeline, running the CDR demo, or troubleshooting the custom consumer. Triggers: kafka consumer, kafka snowflake, snowpipe streaming kafka, CDR demo, kafka to snowflake, custom consumer, streaming ingest kafka.
64
77%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./custom-kafka-consumer/.cortex/skills/custom-kafka-consumer/SKILL.mdQuality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is an excellent skill description that clearly articulates specific capabilities (set up, configure, run, debug), names the precise technology stack (Kafka, Snowflake, Snowpipe Streaming SDK v2), and provides explicit 'Use when' guidance along with a comprehensive list of trigger terms. It follows third-person voice and is concise yet thorough, making it easy for Claude to select this skill accurately from a large pool.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: 'set up, configure, run, and debug a custom Kafka consumer that streams data into Snowflake via Snowpipe Streaming SDK v2.' This clearly names the technology stack and the specific operations supported. | 3 / 3 |
Completeness | Clearly answers both 'what' (set up, configure, run, and debug a custom Kafka consumer streaming to Snowflake via Snowpipe Streaming SDK v2) and 'when' (explicit 'Use when' clause with scenarios plus a dedicated 'Triggers' list). | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms including variations like 'kafka consumer', 'kafka snowflake', 'snowpipe streaming kafka', 'CDR demo', 'kafka to snowflake', 'custom consumer', 'streaming ingest kafka'. These are terms users would naturally use when seeking help with this pipeline. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive niche combining Kafka, Snowflake, Snowpipe Streaming SDK v2, and the CDR demo. Very unlikely to conflict with other skills given the specific technology combination and explicit trigger terms. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
55%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill excels at actionability and workflow clarity — every step is executable with explicit validation checkpoints and error recovery. However, it severely suffers from being a monolithic document (~500+ lines) that inlines massive SQL blocks (semantic views, multiple forecast models, AI verified queries) that should be split into separate reference files. The content is high-quality but poorly structured for context window efficiency.
Suggestions
Extract Steps 7a-7f (ML FORECAST models, semantic view, AI verified queries) into a separate FORECAST_MODELS.md or PREDICTIVE_MAINTENANCE.md file, referenced from the main SKILL.md with a one-line link.
Move the semantic view DDL and AI_VERIFIED_QUERIES block into a dedicated SQL file (e.g., semantic_view.sql) referenced from the skill.
Move the 'Key Consumer Patterns (Reference)' section into a separate CONSUMER_PATTERNS.md file since it's reference material, not workflow steps.
Move the troubleshooting section into a TROUBLESHOOTING.md file and link to it from the main skill with a brief summary of common issues.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely long (~500+ lines) with massive SQL blocks for semantic views, AI verified queries, and multiple forecast models that could easily be split into separate reference files. The semantic view DDL alone is over 100 lines of inline content. While individual sections are reasonably efficient, the overall document is far too large for a SKILL.md, consuming excessive context window. | 1 / 3 |
Actionability | Every step provides fully executable, copy-paste ready commands — bash commands for Kafka setup, complete SQL DDL statements, Maven build/run commands, Java code snippets with exact class paths, and specific configuration file contents. The producer command table and troubleshooting section provide concrete, specific guidance. | 3 / 3 |
Workflow Clarity | The workflow is clearly sequenced across 8 steps with explicit validation checkpoints ('STOP', 'MANDATORY STOPPING POINT'), a data readiness gate before ML training (≥300 rows with a polling query), error recovery guidance in troubleshooting, and a user-confirmation gate before teardown. The feedback loop for the row count check is well-designed. | 3 / 3 |
Progressive Disclosure | This is a monolithic wall of text with no references to external files despite being extremely long. The semantic view DDL (~100 lines), AI verified queries, forecast sub-steps (7a-7f), key consumer patterns, and troubleshooting could all be split into separate reference files. Everything is inlined in a single document with no bundle files to support it. | 1 / 3 |
Total | 8 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
skill_md_line_count | SKILL.md is long (706 lines); consider splitting into references/ and linking | Warning |
Total | 10 / 11 Passed | |
- Commit
25dfdc4
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.