neo4j-kafka-skill
Configure and operate the Neo4j Connector for Kafka (sink + source) and the native Neo4j CDC API. Covers Cypher/Pattern/CUD sink strategies, CDC-based and query-based source, exactly-once semantics, DLQ error handling, Confluent Cloud managed connector, schema registry (Avro/JSON), and native db.cdc.query cursor-loop patterns (Neo4j 5.13+ Enterprise/Aura BC/VDC). Use when streaming Kafka events into Neo4j, streaming Neo4j changes to Kafka, or querying Neo4j change events without Kafka. Does NOT handle Cypher query authoring — use neo4j-cypher-skill. Does NOT handle bulk CSV/file import — use neo4j-import-skill. Does NOT handle GDS algorithms — use neo4j-gds-skill.
72
88%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Quality
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 thoroughly covers specific capabilities, includes rich natural trigger terms, explicitly states both what it does and when to use it, and proactively delineates boundaries with related skills. The negative boundary clauses are a particularly strong feature for disambiguation in a multi-skill environment. The description uses proper third-person voice throughout.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions and strategies: Cypher/Pattern/CUD sink strategies, CDC-based and query-based source, exactly-once semantics, DLQ error handling, Confluent Cloud managed connector, schema registry (Avro/JSON), and native db.cdc.query cursor-loop patterns with version specifics. | 3 / 3 |
Completeness | Clearly answers both 'what' (configure and operate Neo4j Connector for Kafka with detailed strategies and features) and 'when' (explicit 'Use when streaming Kafka events into Neo4j, streaming Neo4j changes to Kafka, or querying Neo4j change events without Kafka'). Also includes explicit negative boundaries directing to other skills. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural terms users would say: 'Kafka', 'Neo4j', 'CDC', 'sink', 'source', 'streaming', 'Confluent Cloud', 'Avro', 'JSON', 'change events', 'DLQ', 'exactly-once'. These are the exact terms a user working in this domain would use. | 3 / 3 |
Distinctiveness Conflict Risk | Extremely distinctive with a clear niche (Neo4j-Kafka integration and CDC). The explicit 'Does NOT handle' clauses with skill redirects (neo4j-cypher-skill, neo4j-import-skill, neo4j-gds-skill) actively prevent conflicts with related skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a strong, highly actionable skill with excellent concrete examples covering all connector strategies, native CDC API, and operational concerns like EOS and DLQ. Its main weakness is length — the SKILL.md tries to be both overview and reference, inlining detailed structures (event schema, selector table) that the referenced files should contain. The workflow clarity and error handling guidance are exemplary.
Suggestions
Move the full event structure JSON and selector reference table to references/cdc-api.md, keeping only a brief summary and link in the main SKILL.md
Move detailed per-strategy config examples to references/sink-config.md, retaining only one representative example (e.g., Cypher strategy) inline with links to the rest
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is generally efficient and avoids explaining basic concepts Claude already knows, but it's quite long (~350 lines). Some sections like the full event structure JSON and the extensive selector reference table could be offloaded to reference files. The decision table and checklist are valuable, but the overall length pushes it toward verbose for a SKILL.md overview. | 2 / 3 |
Actionability | Excellent actionability throughout — every strategy includes complete, copy-paste-ready JSON configs and executable Cypher/Python code. The cursor-loop Python pattern, EOS setup steps, DLQ config, and pattern syntax rules are all concrete and immediately usable. | 3 / 3 |
Workflow Clarity | Multi-step processes are clearly sequenced with explicit validation checkpoints. The EOS setup has numbered steps (create constraint first, then configure), the checklist provides a verification loop, and the common errors table serves as a troubleshooting feedback mechanism. The cursor-loop pattern includes proper cursor advancement logic. | 3 / 3 |
Progressive Disclosure | References to external files (references/sink-config.md, references/cdc-api.md) are present and well-signaled, but no bundle files were provided to confirm they exist. More importantly, the SKILL.md itself contains substantial inline detail (full event structure, complete selector reference table, extensive config examples) that would be better placed in those reference files, making the main file a leaner overview. | 2 / 3 |
Total | 10 / 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 |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
- Repository
- neo4j-contrib/neo4j-skills
- Commit
66ed0e1
- 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.