Content
73%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 strong operational skill: fully sequenced workflows with dry-run/confirm/read-back validation loops for every mutating path, concrete CLI commands with realistic JSON arguments, and appropriate one-level references for platform-specific detail. Its main cost is length — guardrail rules (approval reuse, dry-run-first, idempotency, read-back) are repeated across many sections, and the imported-review-skills discovery flow could live in a reference file.
Suggestions
Consolidate the approval/dry-run/idempotency/read-back rules into a single authoritative section (e.g. the existing Guardrails) and reference it from the workflow steps instead of restating the rules in Instructions, Preflight, the discovery flow, and Preview-and-apply — this would materially shorten the body without losing the constraints.
Move the multi-step 'Discover a source repository' and imported-review-skills configuration detail into a reference file (e.g. references/review-skills-setup.md) with a short inline summary, keeping SKILL.md as an overview with one-level-deep pointers.
Tighten the Quick start by replacing the "..." placeholder arguments (name, category, content, good/bad examples) with one complete worked example of `qodo rules create`, so the primary capture path is copy-paste ready.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and operational with no padding about concepts Claude already knows, but the approval/dry-run/idempotency/read-back rules are restated across Instructions, Preflight, the discovery flow, Preview-and-apply, and Guardrails, and several long caveat passages could be tightened. This fits the 3 anchor (mostly efficient but could be tightened) rather than 2, since the problem is repetition of genuinely necessary rules, not unnecessary explanations or padding. | 3 / 5 |
Actionability | Concrete, mostly copy-paste-ready commands with full JSON payloads (e.g. `qodo rules set-state --args '{"rule_ids":[123,124],"state":"inactive","dry_run":true}'`, bulk previews, `--args-file discovery.json` with its contents shown). Not 5 because several quick-start commands carry "..." placeholders and the skill itself labels them "illustrative, not guaranteed current" — minor gaps, though the live-schema verification rule justifies that flexibility. | 4 / 5 |
Workflow Clarity | Multi-step flows are explicitly sequenced with validation checkpoints and feedback loops for destructive/batch operations: version probe gate → preflight → dry-run → "show the count/blast radius" → confirm → "read the targets again and compare the fields" → "Get every affected id after applying" → reconcile bulk counts, plus an error-handling section mandating read-back before any retry of an uncertain mutation. This matches the 5 anchor (explicit validation steps, error-recovery loops, checklists) — a batch/destructive skill with validation throughout. | 5 / 5 |
Progressive Disclosure | Clear section headers and three real, well-signaled one-level-deep references ([manual-update procedure](references/skill-updates.md), [shell guidance](references/windows-shells.md), [launcher recovery](references/windows-launcher.md)) loaded at the right moments; all referenced paths exist and none nest further. Not 5 because the ~480-line body inlines the entire review-skills discovery/configuration sub-workflow, which is a natural candidate for its own reference file. | 4 / 5 |
Total | 16 / 20 Passed |