Content
85%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
Highly actionable, well-sequenced content with strong validation checkpoints for a database-integration skill. Main weakness is progressive disclosure: everything lives in one ~300-line file with no bundle references, despite editor-specific configs that would split cleanly.
Suggestions
Move per-editor config blocks (Claude Desktop, VS Code, Cursor, Windsurf, Kiro, Cline, Antigravity) into one-level-deep reference files under references/ (e.g. editors/claude-desktop.md) and link from SKILL.md to improve progressive_disclosure.
De-duplicate the repeated JSON env blocks: Claude Desktop and Kiro blocks are near-identical to Claude Code's — reference the shared structure and only show what differs (file path, the ${VARIABLE} syntax).
Consider extracting the HTTP Transport and APOC Requirement sections into a separate reference, keeping SKILL.md as a focused stdio quick-start overview.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Dense and mostly efficient, but full JSON config blocks are repeated for Claude Code, Claude Desktop, and Kiro while other editors get a 'same structure as Claude Code' reference — minor redundancy that could be tightened. Not a 5 because of that duplicated boilerplate; not a 3 because almost everything earns its place. | 4 / 5 |
Actionability | Fully executable throughout: pip/which/cypher-shell commands, complete per-editor JSON configs with real env-var structure, HTTP transport flags, base64 auth generation, and a troubleshooting table with specific fixes. | 5 / 5 |
Workflow Clarity | Clear 5-step sequence (install → credentials → configure → restart → smoke test) with explicit validation checkpoints (connectivity test before config, anti-footgun smoke test, APOC verify, JSON validate), a checklist, and a troubleshooting feedback loop for database operations. | 5 / 5 |
Progressive Disclosure | Well-sectioned with clear headers, but no bundle files exist and ~300 lines are entirely inline; the eight per-editor config blocks and HTTP transport details could be split into one-level-deep reference files. Not a 4 because content that belongs in separate files is inlined; not a 2 because structure and sectioning are genuinely good. | 3 / 5 |
Total | 17 / 20 Passed |