Content
56%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.
The body delivers a well-sequenced, executable workflow with a genuine connection-state feedback loop, but it is dragged down by a large amount of inlined, partly-fabricated padding that wastes tokens and should live in a separate reference file.
Suggestions
Delete the fabricated "Recipe *" entity list (Recipe Machine Learning, Recipe AI, Recipe Disaster Recovery, etc.) which is irrelevant padding unrelated to Square.
Move the full entity list and popular-actions table into a separate reference file and keep SKILL.md as a concise overview that links to it.
Replace the repeated "No description" entries in the popular-actions table with real one-line descriptions so the table is actually actionable.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Nearly half the body is a ~130-item bullet list of entity names (lines 23-154) including fabricated padding like "Recipe Machine Learning", "Recipe Disaster Recovery", and "Recipe Artificial Intelligence" that are irrelevant to using the integration. | 2 / 5 |
Actionability | It provides concrete, executable commands (npm install, membrane login, connection ensure, action list/run, request) with flags, though the popular-actions table repeats "No description" for every row and uses placeholders like CONNECTION_ID. | 4 / 5 |
Workflow Clarity | A clear install -> authenticate -> connect -> poll-for-ready -> search -> run -> proxy sequence exists, with an explicit feedback loop for connection states (READY/BUILDING/CLIENT_ACTION_REQUIRED), but there is no verify/retry guidance for action-run outcomes. | 4 / 5 |
Progressive Disclosure | Section headers give reasonable structure, but there are no reference files at all and the large entity dump plus 20-row action table are inlined content that clearly belongs in a separate reference file. | 3 / 5 |
Total | 13 / 20 Passed |