Content
85%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
A lean, well-structured instruction skill with a clear sequenced workflow and verification step, but its apply-changes step relies on placeholder invocations rather than copy-paste-ready examples.
Suggestions
Add a minimal sample .lsp.json record (or a concrete `/plugin <subcommand> <args>` invocation) so the apply step is copy-paste ready and lifts actionability from 2 to 3.
Show one example of an `lspServers` manifest entry (inline record vs relative file path) to make Step 2's config-source guidance concrete.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The ~40-line body is lean and assumes Claude's competence, explaining only Kode-specific behavior (e.g. what 'LSP enabled' means in Kode) and avoiding general-knowledge padding; every section earns its place. | 3 / 3 |
Actionability | Names concrete commands (/lsp, /plugin via SlashCommand, Edit on .lsp.json) and specific config rules (command must be executable, use args, stdio, avoid restartOnCrash), but the '/plugin ...' ellipsis is a placeholder and no copy-paste-ready example (e.g. a sample .lsp.json record) is given. | 2 / 3 |
Workflow Clarity | Steps 1-4 are clearly sequenced (facts -> config source -> apply -> verify) with an explicit verification checkpoint in Step 4 (re-run /lsp, attempt an LSP tool call) and a decision rule in Step 1. | 3 / 3 |
Progressive Disclosure | A single-purpose skill under 50 lines with no bundle files and well-organized sections (Non-negotiables, definition, Steps 1-4, Notes), which meets the simple-skills allowance for a top score. | 3 / 3 |
Total | 11 / 12 Passed |