Content
85%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.
A well-structured instruction-only skill body: clear sequenced workflow with explicit validation checkpoints, and the large catalog correctly externalized into a single well-signaled reference. Weakest points are repeated truthfulness reminders that could be consolidated and the absence of a small example skeleton for the command packs it mandates.
Suggestions
Consolidate the truthfulness/verified-evidence rules (currently repeated in sections 2, 3, 4, 5, and 6) into a single 'operating rules' statement to reduce token cost.
Add a short skeleton example of one command-pack block (invocation code block + verification terminal block) so the required SPECKITINIT.md format is concrete rather than only described.
Trim meta-instructions like 'Treat it as a reference library, not a document to copy' to a single imperative, since the surrounding bullet list already encodes the curation rules.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Dense, non-obvious domain guidance with no filler explanations of concepts Claude already knows, but the truthfulness/verified-evidence theme recurs across sections 2-6 ('do not claim installation... succeeded', 'verified local evidence', 'Never claim that Spec Kit is installed... without filesystem and CLI evidence') and could be consolidated. Fits anchor 4 ('minor instances of over-explanation that could be trimmed'), not 5. | 4 / 5 |
Actionability | Concrete commands ('specify --help', 'specify integration --help', the official repo URL), named files ('.specify/', 'specs/', 'SPECKITINIT.md'), the four required command packs, and the full ordered sequence (specify, clarify, plan, checklist, tasks, analyze, implement, converge). No example skeleton of a command pack is shown, but deferring exact '/speckit.*' syntax is explicitly justified as version-dependent and sourced from the installed integration, so this is mostly executable guidance with minor gaps (anchor 4). | 4 / 5 |
Workflow Clarity | Clear six-step sequence (discover -> verify CLI -> curate catalog -> author command packs -> constitution/maintenance -> completion standard) with an explicit validation section: dry-run shell validation, link/fence/placeholder checks, diff inspection, and reporting skipped checks and unknowns. Error feedback is present ('If health checks fail, report that result and do not claim installation, upgrade, or integration succeeded'). Matches anchor 5. | 5 / 5 |
Progressive Disclosure | The body is a skimmable overview of rules, and the 44KB workflow catalog is appropriately externalized to a clearly-signaled, verified, one-level-deep reference ('Read the [reusable Spec Kit workflow catalog](references/speckit-skills-guides-workflows/SPECKIT-workflows.md)'). Matches anchor 5: clear overview with well-signaled one-level-deep references. | 5 / 5 |
Total | 18 / 20 Passed |