Content
53%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 is a clean progressive-disclosure pointer: one real, well-structured reference file and an explicit mandate to treat its safety and validation requirements as binding. Its weaknesses are padded trigger phrasing, a duplicated description line, and the absence of any section index or inline critical checks that would make focused loading and standalone use actionable.
Suggestions
Collapse the ten repeated 'User mentions or implies:' lines into a single line ('Use when the user mentions: algolia, instantsearch, typeahead, autocomplete, faceted search, search index, search-as-you-type...') and delete the line that duplicates the frontmatter description.
Add a brief section index of the guide (Patterns, Sharp Edges, Validation Checks, Collaboration) to the body so 'load the relevant sections' names the sections to load.
Surface one or two critical validation checks inline (e.g., never ship the admin API key to the client) so the body is not a pure pointer for the highest-risk pitfalls.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is short, but the phrase 'User mentions or implies:' is repeated verbatim ten times and line 12 restates the frontmatter description verbatim — padding that could be collapsed to a single line. This is 'mostly efficient but includes some unnecessary explanation or could be tightened' rather than score 2, since the total token footprint is still small. | 3 / 5 |
Actionability | The body gives one concrete, verifiable directive ('Read [the detailed guide](references/detailed-guide.md) before executing this skill') plus a focused vs. end-to-end loading strategy, and the referenced file is real and complete. It stays at 3 rather than 4 because the body itself contains no commands, code, or implementation steps — all executable guidance lives one file away. | 3 / 5 |
Workflow Clarity | The reading sequence is unambiguous and validation is explicitly mandated ('Treat its safety, prerequisites, and validation requirements as mandatory'), but the body shows no implementation steps or validation checkpoints of its own, and the skill covers batch-sensitive indexing operations. This matches 'sequence present but checkpoints missing or implicit'. | 3 / 5 |
Progressive Disclosure | A single well-signaled, one-level-deep reference (references/detailed-guide.md, ~25 KB, no nested references) with an appropriate split — the heavy patterns live outside SKILL.md. Kept below 5 because 'For focused work, load the relevant sections' is not supported by any section index in the body, leaving navigation to guesswork. | 4 / 5 |
Total | 13 / 20 Passed |