sw-eval
CommandRun Specwright eval suite. Spawns subagents to run skills in fixture workdirs, then grades results.
Read-only installation health check with repair hints.
$ npx -y skills add Obsidian-Owl/specwright --agent claude-code
description: Read-only installation health check with repair hints.
Use the installed `specwright:sw-doctor` skill for this request. Doctor should report whether runtime roots are `project-visible` under `.specwright-local/` or `git-admin` under `.git/specwright/`, and it should route shipped-state repairs through `/sw-status --repair {unitId}` instead of inventing a separate repair surface.
$ARGUMENTS
Craft quality software with AI discipline. Spec-driven development plugin for Claude Code and Opencode — quality gates, adversarial testing, and evidence capture.
Run Specwright eval suite. Spawns subagents to run skills in fixture workdirs, then grades results.
Explicitly adopt an existing work into the current worktree.
Periodic codebase health check for systemic technical debt.
TDD implementation of one work unit. If branch-head freshness blocks and policy is rebase or merge, recover in-stage; manual is…
Investigation-first debugging. Scope, investigate, diagnose, then fix/log/defer.
Interactive solution architecture with adversarial critique and assumption surfacing.
Implement tasks incrementally — build, test, verify, commit. Add "auto" to run the whole plan in one approved pass.