neo4j-mcp-skill
Use when installing, configuring, or troubleshooting the official Neo4j MCP server (neo4j/mcp): connecting Claude Code, Claude Desktop, Cursor, Windsurf, VS Code, Kiro, or other MCP-compatible editors to a Neo4j database via stdio or HTTP transport. Covers the four MCP tools (get-schema, read-cypher, write-cypher, list-gds-procedures), read-only mode, and multi-database configuration. Does NOT cover writing Cypher queries via those tools — use neo4j-cypher-skill. Does NOT cover agent memory — use neo4j-agent-memory-skill. Does NOT cover Aura instance provisioning — use neo4j-aura-provisioning-skill.
92
88%
Does it follow best practices?
Impact
100%
1.33xAverage score across 3 eval scenarios
Risky
Do not use without reviewing
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 hits all the marks. It opens with an explicit 'Use when' trigger clause, lists specific concrete capabilities and supported tools/editors, and proactively disambiguates itself from three related skills with clear boundary statements. The description is comprehensive yet focused, making it highly effective for skill selection among a large pool of skills.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: installing, configuring, troubleshooting the Neo4j MCP server, connecting various editors, covers specific MCP tools (get-schema, read-cypher, write-cypher, list-gds-procedures), read-only mode, multi-database configuration, and stdio/HTTP transport. | 3 / 3 |
Completeness | Clearly answers both 'what' (installing, configuring, troubleshooting the Neo4j MCP server, its four tools, read-only mode, multi-database config) and 'when' (opens with 'Use when installing, configuring, or troubleshooting...'). Also explicitly defines boundaries with 'Does NOT cover' clauses directing to other skills. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural terms users would say: 'Neo4j MCP server', 'neo4j/mcp', 'Claude Code', 'Claude Desktop', 'Cursor', 'Windsurf', 'VS Code', 'Kiro', 'stdio', 'HTTP transport', 'get-schema', 'read-cypher', 'write-cypher', 'read-only mode', 'multi-database'. These are exactly the terms a user would mention when needing help with MCP server setup. | 3 / 3 |
Distinctiveness Conflict Risk | Exceptionally distinctive — explicitly delineates boundaries against three related skills (neo4j-cypher-skill, neo4j-agent-memory-skill, neo4j-aura-provisioning-skill) with 'Does NOT cover' clauses, making it very clear when this skill should and should not be selected. | 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 high-quality, highly actionable skill that provides concrete, copy-paste-ready configuration for multiple editors with clear validation steps and excellent troubleshooting guidance. Its main weakness is length — the repeated JSON blocks for similar editors and some over-explanation inflate the token cost. The content would benefit from factoring out repetitive editor configs into a reference file or collapsing similar editors into a single template with a path table.
Suggestions
Consolidate the editor config blocks: show one canonical JSON template, then provide a table mapping editor → config file path → key name (`mcpServers` vs `servers`), rather than repeating near-identical JSON 6+ times.
Consider splitting HTTP transport, APOC requirements, and the troubleshooting table into a separate reference file to keep the main SKILL.md focused on the core stdio setup workflow.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is mostly efficient and well-targeted, but there's some verbosity — the absolute path explanation is over-explained, multiple editor configs repeat nearly identical JSON blocks (Claude Code, Claude Desktop, Cursor, Windsurf all share the same structure), and some notes could be tightened. However, the editor-specific path differences do add genuine value. | 2 / 3 |
Actionability | Excellent actionability throughout: every step has concrete, copy-paste-ready commands and JSON configs. Installation, credential verification, editor configuration, smoke testing, and troubleshooting all provide specific executable guidance with real commands and expected outputs. | 3 / 3 |
Workflow Clarity | The 5-step installation workflow is clearly sequenced with explicit validation checkpoints — Step 2 verifies connectivity before configuring the editor, Step 5 provides a specific smoke test with expected output and warns against a common false-positive test. The troubleshooting table and final checklist provide additional verification. The feedback loop of 'verify before configuring' prevents wasted debugging. | 3 / 3 |
Progressive Disclosure | The content is well-structured with clear sections and a logical flow, but it's a long monolithic file (~250 lines of substantive content). The repeated editor config blocks could be collapsed or moved to a separate reference file. HTTP transport and APOC sections could also be split out. No bundle files are provided to offload detail, so everything is inline. | 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 |
|---|---|---|
allowed_tools_field | 'allowed-tools' contains unusual tool name(s) | 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.