Content
87%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.
A tight, highly actionable single-file skill with concrete code and templates. Its main gap is the absence of an explicit verification checkpoint for a batch export, which keeps workflow clarity at 2.
Suggestions
Add a verification step after the write (e.g., confirm the file exists and that the number of blocks written equals the number of memories fetched) to turn the batch export into a validate-loop workflow.
Specify what to do on partial pagination failure (e.g., mid-sequence error) so the error-handling section covers the batch path, not just the empty/error and write-failure cases.
Clarify the pagination loop exit condition explicitly (e.g., 'stop when a page returns fewer than page_size results and no next cursor') rather than leaving it implicit.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and efficient, providing only operational detail (exact filters, format templates, filename pattern) without explaining concepts Claude already knows, matching the top anchor. | 3 / 3 |
Actionability | It gives a concrete get_memories call with exact filters, an exact YAML-frontmatter block template, a precise output filename pattern, and copy-paste-ready summary text, matching the fully-executable anchor. | 3 / 3 |
Workflow Clarity | The five steps are clearly sequenced with error handling, but this batch operation (exporting all memories) lacks an explicit validation/verification checkpoint confirming the count reconciles or the file was written correctly, which caps the score at 2 per the rubric's batch-operation guideline. | 2 / 3 |
Progressive Disclosure | It is a single, well-organized SKILL.md with clear sections (Execution steps, Error Handling) and no nested external references, which is appropriate for a self-contained single-purpose skill. | 3 / 3 |
Total | 11 / 12 Passed |