Content
27%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill provides comprehensive coverage of Supabase tools with well-structured workflow sequences and useful pitfall warnings. However, it is severely bloated with duplicated content (pitfalls repeated across sections), excessive parameter documentation that could be discovered via RUBE_SEARCH_TOOLS, and appears to contain corrupted embedded content. It lacks executable examples and validation checkpoints for destructive operations.
Suggestions
Reduce content by 50%+: remove the duplicated 'Known Pitfalls' section (already covered per-workflow), trim parameter lists to only non-obvious details, and remove PostgREST operator lists that Claude already knows or can discover via RUBE_SEARCH_TOOLS.
Add concrete example tool call payloads showing input parameters and expected output structure for at least the SELECT_FROM_TABLE and RUN_SQL_QUERY workflows.
Add explicit validation steps for write operations: after RUN_SQL_QUERY for INSERT/UPDATE/DELETE, include a verification SELECT to confirm the mutation succeeded, with error recovery guidance.
Split detailed parameter references and the quick reference table into a separate REFERENCE.md file, keeping SKILL.md as a concise overview with workflow sequences and links to details.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is extremely verbose and repetitive. The 'Known Pitfalls' section largely duplicates pitfalls already listed under each workflow. The 'Common Patterns' section repeats ID resolution steps already covered. PostgREST filter operators and many parameter details are things Claude already knows or could discover via RUBE_SEARCH_TOOLS. The content also appears to contain corrupted/duplicated rubric text embedded within it, inflating length enormously. | 1 / 3 |
Actionability | The skill provides specific tool names, parameter names, and key parameter details, which is concrete guidance. However, there are no executable code examples, no example tool call payloads, and no example inputs/outputs showing what a successful query looks like. The guidance describes what to do but doesn't show copy-paste-ready tool invocations. | 2 / 3 |
Workflow Clarity | Workflows are clearly sequenced with numbered steps and labeled prerequisites/required/optional annotations. However, for database write operations (INSERT/UPDATE/DELETE via RUN_SQL_QUERY), there are no explicit validation checkpoints or feedback loops — no 'verify the mutation succeeded' step or error recovery guidance. Per rubric rules, missing feedback loops for database operations caps this at 2. | 2 / 3 |
Progressive Disclosure | This is a monolithic wall of text with no bundle files and no references to external documents. All detailed parameter lists, pitfalls, and the full quick reference table are inlined in a single massive file. The extensive parameter documentation and repeated pitfalls sections should be split into separate reference files. | 1 / 3 |
Total | 6 / 12 Passed |