Content
46%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill provides a reasonable high-level framework for cross-jurisdictional legal referencing with a clear workflow structure and CLI-based interface. However, it suffers from unverifiable CLI commands with no expected outputs, missing validation/error-handling steps for a high-stakes legal domain, and references to templates and bundle files that don't exist. The content is moderately verbose with some unnecessary framing that could be trimmed.
Suggestions
Add expected output examples for each CLI command (e.g., show what `lex search` returns) to make the skill truly actionable and verifiable.
Add explicit validation and error-handling steps: what happens if `lex verify` returns no sources, or if a jurisdiction is partially covered? Include a feedback loop for draft validation.
Provide the referenced bundle files (e.g., `templates/02_employment_workforce.md`) or restructure references to clearly indicate what supporting materials exist and where they live.
Trim the Overview and 'When to Use This Skill' sections — Claude doesn't need to be told what a 'truth engine' is or when to use cross-referencing; focus on the actionable workflow and constraints.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill includes some unnecessary explanation (e.g., the Overview section restates what LEX is in a way Claude doesn't need, and 'When to Use This Skill' is somewhat verbose). However, the workflow steps and examples are reasonably tight. There's room to trim the introductory and best-practices sections. | 3 / 5 |
Actionability | The skill provides CLI commands like `lex search`, `lex get`, `lex draft`, and `lex verify`, which appear concrete. However, these commands are not verifiable — there's no indication of how the `lex` CLI is installed, configured, or what its actual output looks like. The examples show bash commands but no expected output, making it unclear whether these are real executable tools or aspirational pseudocode. | 3 / 5 |
Workflow Clarity | The four-step workflow (Identify Jurisdiction → Search → Draft → Verify) provides a clear sequence, but there are no validation checkpoints or feedback loops. Step 4 ('Verify Authority') is a verification step but there's no guidance on what to do if verification fails or if sources are missing. For a skill involving legal document generation (a high-stakes domain), the absence of explicit error handling and validation loops is a notable gap. | 3 / 5 |
Progressive Disclosure | The skill references template paths like `templates/02_employment_workforce.md` and related skills like `@employment-contract-templates`, but no bundle files are provided. There's no actual supporting documentation to navigate to. The content that could benefit from separate files (e.g., jurisdiction coverage lists, template catalogs) is neither inlined nor referenced in a structured way. | 2 / 5 |
Total | 11 / 20 Passed |