Content
100%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.
This is a model lean single-purpose skill: complete and unambiguous instructions in two sentences with no padding, no over-explanation, and nothing that needs splitting into bundle files. Both the tool invocation and the exact output formatting are fully specified.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Two sentences with zero padding: "Call the ctx_stats tool and show the text block of its reply to the user unchanged, inside a code block so the columns stay aligned" delivers tool, transformation, format, and rationale; every token earns its place, matching the lean anchor 5. | 5 / 5 |
Actionability | "Call the ctx_stats tool", "show the text block of its reply to the user unchanged, inside a code block", and "Add one sentence under it that names the largest saving" are fully executable instructions covering the single common case; per the rubric's code-vs-instruction note, the absence of code is not penalized when guidance is this specific. | 5 / 5 |
Workflow Clarity | The single-task sequence (call tool, render output unchanged in a code block, add the one-sentence summary naming the largest saving) is unambiguous, and as a read-only display task no validation checkpoints are needed — the simple-skill exception applies. | 5 / 5 |
Progressive Disclosure | The skill is two sentences with no external material needed and no bundle files exist, so everything appropriately lives inline in SKILL.md; the under-50-line simple-skill exception for progressive disclosure applies. | 5 / 5 |
Total | 20 / 20 Passed |