sw-eval
01Command Claude Code
Run Specwright eval suite. Spawns subagents to run skills in fixture workdirs, then grades results.
not rated 10 5mo ago C 22 tokens
original MIT
specwright
02Plugin Claude Code
Spec-driven development with quality gates. Ensures you get what you asked for.
not rated 10 5mo ago A
tokens not measured
original MIT
sw-adopt
03Command Claude Code
Explicitly adopt an existing work into the current worktree.
not rated 10 5mo ago A 12 tokens
original MIT
sw-audit
04Command Claude Code
Periodic codebase health check for systemic technical debt.
not rated 10 5mo ago A 10 tokens
original MIT
sw-build
05Command Claude Code
TDD implementation of one work unit. If branch-head freshness blocks and policy is rebase or merge, recover in-stage; manual is the fallback.
not rated 10 5mo ago A 30 tokens
original MIT
sw-debug
06Command Claude Code
Investigation-first debugging. Scope, investigate, diagnose, then fix/log/defer.
not rated 10 5mo ago A 17 tokens
original MIT
sw-design
07Command Claude Code
Interactive solution architecture with adversarial critique and assumption surfacing.
not rated 10 5mo ago A 12 tokens
original MIT
sw-doctor
08Command Claude Code
Read-only installation health check with repair hints.
not rated 10 5mo ago A 9 tokens
original MIT
sw-guard
09Command Claude Code
Detect stack and configure deterministic guardrails.
not rated 10 5mo ago A 8 tokens
original MIT
sw-init
10Command Claude Code
Project setup. Creates constitution and charter. Configures gates.
not rated 10 5mo ago A 12 tokens
original MIT
sw-learn
11Command Claude Code
Post-ship capture of patterns and learnings.
not rated 10 5mo ago A 10 tokens
original MIT
sw-pivot
12Command Claude Code
Research-backed rebaselining for planning, building, or verifying work. Preserves completed and shipped scope while revising remaining design, plan, and in-progress work.
not rated 10 5mo ago A 34 tokens
original MIT
sw-plan
13Command Claude Code
Decompose design into work units with testable specs.
not rated 10 5mo ago A 11 tokens
original MIT
sw-research
14Command Claude Code
Deep outward-facing research. External docs, APIs, patterns, and validation.
not rated 10 5mo ago A 15 tokens
original MIT
sw-review
15Command Claude Code
PR comment review. Group by status and respond inline.
not rated 10 5mo ago A 11 tokens
original MIT
sw-ship
16Command Claude Code
Strategy-aware merge via pull request. If branch-head freshness blocks and policy is rebase or merge, recover in the same run; manual is the fallback and still reruns /sw-verify before /sw-ship.
not rated 10 5mo ago A 44 tokens
original MIT
sw-status
17Command Claude Code
Show current Specwright state and progress.
not rated 10 5mo ago A 8 tokens
original MIT
sw-sync
18Command Claude Code
Git housekeeping. Fetch, prune stale branches, and sync with remote.
not rated 10 5mo ago A 14 tokens
original MIT
sw-verify
19Command Claude Code
Interactive quality gates. Shows findings and validates against the spec. If branch-head freshness blocks and policy is rebase or merge, recover in the same run; manual is the fallback and stays on /sw-verify.
not rated 10 5mo ago A 43 tokens
original MIT
Agent Claude Code
Strategic architecture advisor. Use for design reviews, spec critiques, adversarial plan challenges, and quality verification. READ-ONLY.
not rated 10 5mo ago A 31 tokens
original MIT
Agent Claude Code
Fixes build and test failures with minimal changes. Gets the build green quickly without architectural changes or refactoring.
not rated 10 5mo ago A 28 tokens
original MIT
Agent Claude Code
Focused task executor for TDD implementation. Builds exactly one work unit at a time. Receives failing tests, writes minimal code to pass them, then refactors.
not rated 10 5mo ago A 38 tokens
original MIT
Agent Claude Code
Integration test engineer for non-unit tiers. Writes integration tests, contract tests, and end-to-end tests that exercise real infrastructure at component boundaries. Never writes skip conditions for missing infrastructure.
not rated 10 5mo ago A 43 tokens
original MIT
Agent Claude Code
Documentation and reference researcher. Fetches official docs, verifies technical information, and summarizes findings. READ-ONLY.
not rated 10 5mo ago A 28 tokens
original MIT
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: