Content
35%Weight 40%Scale 1-3Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body has the right lightweight shape and a clean section structure, but it is content-thin: no executable guidance, generic placeholder instructions, and a reference to a resource file that is absent. The overview is well-intentioned but currently non-functional for actual projection work.
Suggestions
Replace the generic "Instructions" platitudes with a concrete projection-building workflow, e.g. an explicit sequence of subscribing to the event stream, applying handlers per event type, idempotently updating the read store, and a verification/replay checkpoint.
Add at least one executable example (e.g. a small code or pseudo-schema snippet showing an event handler updating a read model) so the skill is actionable rather than descriptive.
Create the referenced `resources/implementation-playbook.md` (or change the pointer to an existing file) — the body cites it twice but no `resources/` directory exists, leaving a broken reference.
Remove the tautological "Do not use this skill when" entries ("The task is unrelated to projection patterns") which add no information.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is short and avoids explaining concepts Claude already knows, but the generic "Do not use this skill when" bullets and platitudinous "Instructions" ("Apply relevant best practices", "Provide actionable steps") are filler that could be tightened, so it sits at level 2 rather than the every-token-earns-its-place level 3. | 2 / 3 |
Actionability | There is no concrete code, command, or worked example; guidance like "Apply relevant best practices and validate outcomes" is vague and describes rather than instructs, matching the level-1 anchor. The only concrete pointer is to a resource file that does not exist. | 1 / 3 |
Workflow Clarity | A nominal sequence is present (clarify goals → apply practices → validate outcomes), but these are generic platitudes with no real projection-building steps or explicit validation checkpoints, fitting the level-2 anchor of a sequence with missing/implicit checkpoints rather than a concrete workflow. | 2 / 3 |
Progressive Disclosure | The file is structured as an overview pointing to one detailed reference, which is the right shape, but the referenced `resources/implementation-playbook.md` does not exist on disk, so navigation is broken — level 2 rather than 3 because the clearly-signaled reference is a dead path. | 2 / 3 |
Total | 7 / 12 Passed |