Content
75%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 strong procedural skill body: fully sequenced, concrete, and validation-rich, with an exact output template. Remaining room is small — trimming Guardrails redundancy, adding a recovery path for archive-name collisions, and specifying the date-generation command.
Suggestions
Add an error-recovery loop to step 5: after a collision failure, instruct to prompt for an alternative name or date suffix and retry the mv rather than just failing.
Give the concrete date command (e.g. `date +%F`) for building the `YYYY-MM-DD-<change-name>` archive name.
Trim Guardrail entries that verbatim restate steps 1 and 4, keeping only the non-redundant guards.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and assumes Claude's competence (no concept explanations, exact commands), but the Guardrails section restates steps 1 and 4 ("Always prompt for change selection if not provided", "If delta specs exist, always run the sync assessment") — a minor trim candidate that keeps it below anchor 5. | 4 / 5 |
Actionability | Concrete, executable commands are given throughout (`openspec list --json`, `openspec status --change "<name>" --json`, `mkdir -p openspec/changes/archive`, the `mv` invocation), plus verbatim prompt options and an output template; minor gaps (no command for the YYYY-MM-DD date suffix, and the delta-spec comparison method is left unspecified) hold it at 'minor gaps' rather than fully copy-paste ready. | 4 / 5 |
Workflow Clarity | A clearly sequenced 6-step workflow with explicit validation checkpoints (artifact status check, incomplete-task count, sync-state assessment, archive-collision check) and user confirmations via AskUserQuestion; validation is present so the destructive-operation cap does not apply, but the collision path fails with suggestions and no retry loop, which anchor 5's 'feedback loops for error recovery' would require. | 4 / 5 |
Progressive Disclosure | No bundle files exist and the body is well-sectioned (Input, Steps, Output On Success, Guardrails) with sync detail appropriately delegated to the openspec-sync-specs skill; at ~115 lines with everything inline it is well-placed content rather than an ideal split-free overview, fitting anchor 4's 'good structure, minor organization gaps'. | 4 / 5 |
Total | 16 / 20 Passed |