Content
100%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.
An exemplary instruction-only SKILL.md: token-efficient, parameter-specific, and tightly sequenced with explicit guardrails for a risky browser-automation workflow. It earns top marks across all content dimensions without any bundle files.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Lean and instruction-dense with no padding or explanations of concepts Claude already knows; every line delivers actionable tool guidance and assumes competence, matching the score-3 anchor. | 3 / 3 |
Actionability | Concrete, executable guidance throughout: named tools with specific parameters (research(action="fetch-url"), browser_snapshot with interactive: false, browser_capture_secret via ref or redactedKey, data vs resolveData), which for an instruction-only skill is copy-ready direction. | 3 / 3 |
Workflow Clarity | A clear numbered Default Procedure (1-5) with explicit validation checkpoints — 'not completion' criteria, refresh snapshot before acting, inspect state after navigation — and a defined Closeout, giving a well-sequenced flow with feedback guardrails. | 3 / 3 |
Progressive Disclosure | No bundle files are provided and the body is a compact single-purpose skill organized into clearly headed sections (Default Procedure, Secrets, Browser Discipline, Closeout), which per the simple-skills note earns a 3 with well-organized sections alone. | 3 / 3 |
Total | 12 / 12 Passed |