Content
57%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The content is well-organized and points to a shared prerequisite cleanly, with usable discovery commands. It is held back by verbose API prose, the absence of a concrete worked example, and no validation/retry guidance for batch operations.
Suggestions
Trim the `batchUpdate` description to the essential behavior, dropping the replies/ordering prose that Claude can infer from `gws schema`.
Add at least one complete executable example, e.g. `gws slides presentations create --params '{"title":"My Deck"}'`.
Add a validation/verification step for batch operations, such as inspecting the returned replies or re-running `gws slides presentations get` to confirm changes applied.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly efficient and well-structured, but the `batchUpdate` entry is padded with API-doc prose ("Some requests have replies... Other requests do not need to return information... The order of replies matches that of the requests") that could be tightened. | 2 / 3 |
Actionability | Concrete commands are present (`gws slides --help`, `gws schema slides.<resource>.<method>`) and the base invocation syntax is given, but there is no complete worked example with real `--params`/`--json` values that is copy-paste ready. | 2 / 3 |
Workflow Clarity | There is a sensible inspect-then-call sequence (prerequisite → help → schema → build flags → call), but no validation checkpoint or fix-retry loop is given for the batch `batchUpdate` operation, which caps workflow clarity at 2. | 2 / 3 |
Progressive Disclosure | The skill is short and well-organized into clear sections, with a single one-level-deep prerequisite reference to `../gws-shared/SKILL.md` that is clearly signaled; no nested or multi-level references exist. | 3 / 3 |
Total | 9 / 12 Passed |