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-sequenced body with strong validation feedback loops for destructive operations; its main weakness is redundant trigger/tool-selection sections and a monolithic single-file structure that could benefit from splitting out the API reference.
Suggestions
Collapse the overlapping Activation Contract, 'When to use this skill', and 'How to use this skill' sections into a single trigger + tool-selection block to remove redundancy.
Move the detailed 'Available MCP tools' reference into a separate references/ file, leaving SKILL.md as a concise overview that points to it.
De-duplicate the tool-picking guidance that appears both in 'How to use this skill' step 2 and in the per-tool sections.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly efficient, but the Activation Contract, 'When to use this skill', and 'How to use this skill' sections overlap noticeably, and tool-selection guidance is restated later in 'Available MCP tools', so the body could be tightened. | 3 / 5 |
Actionability | Provides concrete tool names, action parameters, call shapes, and copy-pasteable JSON/SQL examples (runQuery payload, _openid column DDL, queryPermissions/managePermissions shapes) covering the common cases. | 5 / 5 |
Workflow Clarity | Five sequenced scenarios with explicit checkpoints: provisioning polls until READY, and the destructive destroy flow requires confirmation, verifies via getInstanceInfo, and recovers from failed tasks via describeTaskStatus. | 5 / 5 |
Progressive Disclosure | Sibling-skill references are one-level-deep and clearly signaled, and sections are well organized, but the full tool reference and all lifecycle flows are inlined in one ~200-line file rather than split into separate bundle files. | 4 / 5 |
Total | 17 / 20 Passed |