By Akhiranandha
Spec-Driven Development cycle for Claude Code without automated tests: spec → build → validate → ship, with iterative updates. Six self-contained skills (one per phase, plus update and run helpers) plus a read-only /spec-lean:status command. Story-driven (US-N IDs), with optional Done-when checks per story for /spec-lean:validate. The ship phase dispatches commit-commands and code-review (from claude-plugins-official); all other phases are entirely self-contained.
Phase 2 of the SDD cycle. Use when the user invokes /spec-lean:build <feature> to implement the spec at docs/specs/<feature>/spec.md story-by-story. Performs git pre-checks (offers git init if missing), scaffolds .gitignore for the language detected, implements bottom-up per User Story, writes per-feature artifacts to docs/specs/<feature>/spec-status.md, updates project-wide docs/codebase-map.md, and prompts a recommended-keep-branch rollback path on hard failure.
Orchestrator for the full SDD cycle. Use when the user invokes /spec-lean:run <feature> to chain spec → build → validate → ship end-to-end. Pauses for explicit user confirmation between every phase (y / edit / stop). Resumable — detects existing artifacts (spec.md, spec-status.md) and offers to skip already-completed phases.
Phase 4 of the SDD cycle. Use when the user invokes /spec-lean:ship <feature> to review, commit, push, PR, and merge. Performs an SDD-aware pre-commit code review of the local diff (BLOCK/CAUTION/GO verdict) — checking story deviations against US-N IDs in spec-status.md, runtime-artifact leaks, and the standard security / data-loss / concurrency issues. Only on GO does it dispatch commit-commands:commit (no remote) or commit-commands:commit-push-pr (with remote), then runs the marketplace /code-review on the PR diff as a second pass before asking for explicit merge permission.
Phase 1 of the SDD cycle. Use when the user invokes /spec-lean:spec <feature> to interactively author a feature spec at docs/specs/<feature>/spec.md (amendments to an existing spec go through /spec-lean:update instead). Produces the SDD 6-section template (Goal, Requirements, User Stories with US-N IDs and optional Done-when checks, Technical details, Out of scope, Edge cases). Sub-capabilities of one feature go in one spec as separate US-IDs — never multiple specs. Pauses at a checkpoint so the user can review before implementation.
Iteration handler for the SDD cycle. Use when the user invokes /spec-lean:update <feature> to amend an existing spec mid-cycle. Snapshots the current spec to spec.md.prev, lets the user revise interactively, diffs the two versions to identify changed US-N IDs, and marks affected stories as in-progress in spec-status.md so the next /spec-lean:build only redoes what changed.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
A Claude Code plugin marketplace by Akhiranandha.
/plugin marketplace add Akhiranandha/custom-claude-plugins
Then install individual plugins:
/plugin install spec-tests-first@akhira-plugins
/plugin install spec-lean@akhira-plugins
/plugin install review-code@akhira-plugins
npx claudepluginhub akhiranandha/custom-claude-plugins --plugin spec-leanTwo-pass code review for Claude Code: dispatches code-quality-reviewer (SonarQube-style maintainability) and security-reviewer (OWASP / CWE-mapped) in parallel, then code-reporter aggregates both into a single timestamped Markdown report under ./reports/. Read-only — never modifies code.
Spec-Driven Development cycle for Claude Code (v2.2.1): /spec-tests-first:init → spec → build → review → fix → validate → ship. v2.2.1 is a documentation/wording patch — 13 audit findings addressed (README skill count, /spec-tests-first:ship Phase 6 ordering wording, /spec-tests-first:run resumability for init, /spec-tests-first:fix sub-label cleanup, /spec-tests-first:build's v1 spec-status.md backward-compat insert in Step 5c, --report flag parsing, etc.). No behavioral changes from v2.2. New /spec-tests-first:init pre-cycle skill auto-detects three cases (fresh repo / existing codebase / migrate flat specs) so STF can be adopted on existing projects, not just greenfield. Per-AC red-green-refactor, in-cycle code-review + fix phases, per-phase status tracking, user-approval gate, scope check, stack-aware test-layout profiles with monorepo support. Zero external plugin dependencies.
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Develop, test, build, and deploy Godot 4.x games with Claude Code. Includes GdUnit4 testing, web/desktop exports, CI/CD pipelines, and deployment to Vercel/GitHub Pages/itch.io.
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.
Comprehensive PR review agents specializing in comments, tests, error handling, type design, code quality, and code simplification
Harness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Access thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.