Content
81%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.
A highly actionable, well-structured overview skill: concrete commands and configs cover the common setup paths, workflows are sequenced with explicit verification steps and error-recovery loops, and deep material is appropriately delegated to six one-level-deep reference files. The main weakness is conciseness — the same compatibility warnings, region lists, and catalog caveats are repeated across sections, and the inline neon.ts detail inflates the overview.
Suggestions
Deduplicate the Data API/PostgREST compatibility caveat (currently stated in at least six sections) into one authoritative statement plus the existing routing table row, and cross-link it instead of restating it.
Consolidate the repeated region-availability list and the neon-auth-not-in-CLI-catalog warning (each stated twice) into single locations referenced from the other mentions.
Move the deep neon.ts material (Function env key semantics, branch callback configuration, Data API type-error walkthrough) into a reference file, keeping a compact overview and one minimal example in SKILL.md.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly efficient — commands, flags, and tables rather than concept explanations — but the same caveats are restated many times: the Data API/PostgREST compatibility rule appears in at least six sections ('Backend Primitives', 'Architecture', 'Convert an app onto Neon', 'Choosing the Right Skill', 'neon init --services', 'Starting without a Neon account'), the region list and the neon-auth-not-in-CLI-catalog warning each appear twice, and time-sensitive values ('72 hours today', '15 minutes today', 'current releases') are inline rather than isolated. This is more than the 'minor instances of over-explanation' of the 4 anchor but well short of the severely padded 2 anchor. | 3 / 5 |
Actionability | Fully executable, copy-paste-ready guidance: exact bash commands with flags ('neon init --agent cursor --org-id <org-id> --project-id <project-id> -y', 'neon mcp --oauth --project --agent <agent> -y'), complete neon.ts TypeScript examples, concrete doc-fetch mechanics ('Append .md to the URL', 'curl -H "Accept: text/markdown"'), and precise operational rules ('Never coerce a missing process.env value to an empty string: that uploads "" and deletes the live key'). | 5 / 5 |
Workflow Clarity | Clear sequenced workflows with explicit validation and feedback loops: the 5-step 'Convert an app onto Neon' checklist ends with 'Verify the app flow (sign-in, upload, API call), not only that env vars landed'; 'neon config plan' provides a dry-run before apply; env pull 'fails fast if the branch is missing a declared service, pointing you at neon deploy'; the type-error section shows the error message and both valid fixes; and auth failures route to 'stop and ask the user to sign in' rather than a substitute path. | 5 / 5 |
Progressive Disclosure | Good structure with well-signaled one-level-deep references — auth, claimable-neon, function-triggers, parse-env, logs-loki, and sdk are each real files under references/ linked from the body ('CLI, MCP, REST, inherited-trigger behavior, and parsers: references/function-triggers.md'), and the reference files themselves are leaves. Falls short of 5 because the ~130-line inline neon.ts deep-dive (Function env semantics, branch callback config, Data API type-error walkthrough) and the duplicated warnings across sections could largely live in a reference, keeping the overview leaner. | 4 / 5 |
Total | 17 / 20 Passed |