By nklisch
Software development workflow skills — design features, implement from plans, refactor codebases, extract patterns, design tests, and ship releases.
Find beautiful code abstractions and cross-cutting simplifications. Applies conceptual lenses (elimination, unification, inversion) to surface bold architectural ideas, then converges into an implement-ready design document.
Sweep a codebase for AI-accumulated cruft (dead code, stale comments, compatibility shims, defensive bloat, over-abstraction), triage findings by confidence tier, then orchestrate parallel cleanup agents with worktree isolation. Use when user says "clean cruft", "dead code", "unused code", "stale comments", or "AI leftovers".
Architectural principles (Ports & Adapters, Single Source of Truth, Generated Contracts). Auto-loads when designing new modules, defining interfaces between layers, discussing system architecture, drawing system boundaries, reviewing architectural decisions, or any time the design or implement skill is active.
Create a detailed design document with typed implementation units. Use when a feature spans multiple files or modules, when scope is unclear before coding, when the user provides a vision or roadmap and wants a plan before implementation, or when implementing would require too many upfront decisions to make inline. Skip if the task is small enough to implement directly.
Design a comprehensive end-to-end test suite for a project. Explores the codebase, understands user-facing behavior from docs and code, then designs two test sets: golden-path user journey tests and adversarial/failure-mode tests. Interactive — asks the user about expected failure behavior before finalizing. Use when starting e2e testing, expanding test coverage, or designing integration tests for a project.
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 software development workflow suite for Claude Code — from ideation through implementation, refactoring, testing, and release.
Available as a Claude Code plugin and via skilltap.
# Step 1: Add the marketplace
/plugin marketplace add nklisch/skills
# Step 2: Install plugins
/plugin install workflow@nklisch-skills
/plugin install skill-authoring@nklisch-skills
Two plugins available:
# Add this tap
skilltap tap add nklisch https://github.com/nklisch/skills
# Install individual skills
skilltap install design
skilltap install implement
skilltap install research
These skills form an iterative pipeline. See docs/workflow-guide.md for the full usage guide.
ideate → roadmap ← project start
↓
design → implement/implement-orchestrator ← core phases (from roadmap)
↓
refactor-design → implement → extract-patterns ← every 2-4 phases
↓
bold-refactor / perf-design → implement ← targeted improvements
↓
expand ← major scope changes (updates foundation docs)
↓
stylistic/structural-refactor-creator ← every 3-5 phases
↓
security-review → design → implement ← security audit pass
↓
feature → design → implement ← quick one-offs near the end
↓
release ← ship it
| Skill | What it does |
|---|---|
| ideate | Interactive project definition workshop. Produces foundation docs (VISION.md, SPEC.md, ARCHITECTURE.md). Run once at project start. |
| roadmap | Generate an agent-optimized project roadmap from foundation docs. Decomposes into phases with dependencies and test checkpoints. |
| expand | Expand project scope — new subsystem, major capability, architectural shift. Updates foundation docs and roadmap. |
| feature | Scope a quick extension or one-off outside the core roadmap. Produces a feature brief that design consumes. |
| design | Produce detailed implementation units with exact interfaces, types, and acceptance criteria. |
| implement | Write code from a design or plan. Single Sonnet agent. Best for <20 files. |
| implement-orchestrator | Opus orchestrator that spawns Sonnet agents. For 20+ files or independent subsystems. |
| research | Investigate libraries, APIs, and SDKs. Produces a research doc and an auto-loading reference skill. |
| refactor-design | Find duplication and missing abstractions. Produces a refactor plan that implement executes. |
| bold-refactor | Find beautiful code abstractions and cross-cutting simplifications using conceptual lenses. Produces an implement-ready design. |
| extract-patterns | Document reusable code patterns for consistency across future work. |
| stylistic-refactor-creator | Interview-based. Generates a project-specific stylistic-refactor skill. |
| structural-refactor-creator | Interview-based. Generates a project-specific structural-refactor skill. |
| perf-design | Profile performance bottlenecks and design optimized solutions. Follows a strict optimization hierarchy. |
| security-review | Comprehensive security audit. Discovers stack, user selects focus domains, produces scored report with severity-classified findings. |
| cruft-cleaner | Sweep for AI-accumulated cruft (dead code, stale comments, defensive bloat). Triages by confidence, parallel cleanup. |
| test-quality | Spec-driven test gap analysis. Derives tests from contracts, not code. Writes the tests. |
| e2e-test-design | Design golden-path and adversarial e2e test suites. Interactive. |
| update-documentation | Align all docs to code after changes. |
| release | Draft changelog, confirm with user, run the project's release mechanism. |
| Skill | What it does |
|---|---|
| write-tool-skill | Create distributable reference skills for your project's tool, CLI, MCP server, or library — for others to install. |
| skill-idea-refiner | Refine a rough skill idea into a well-designed skill. Guides through ideation, scoping, and scaffolding. |
| skill-evaluator | Evaluate skills against type-specific quality rubrics. Produces scored reports with improvements. |
| tool-evaluator | Self-evaluate agent tool usage in the current conversation. Produces a report with recommendations for tool authors. |
Auto-load during design and implementation to enforce standards.
npx claudepluginhub nklisch/skills --plugin workflowDelegate arbitrary task work from Claude Code to a peer local coding agent (Codex, Claude, Gemini)
Delegate arbitrary task work from Claude Code to a peer local coding agent (Codex, Claude, Gemini)
Runtime debugging for AI agents — debug 10 languages via DAP, record and control Chrome via CDP.
End-to-end development workflow: design → draft-plan → orchestrate → review → pr-create → pr-review → pr-merge
Essential development commands for coding, debugging, testing, optimization, and documentation
Focused agentic engineering workflow: design-doc, spec, plan, implement, tdd, refactor, review, address-pr-feedback, browser-verify, explain-visually, compress, branch, and commit.
Software engineering workflows with skills for planning, implementation, quality review, and structured thinking, plus a suite of specialist agents
Helder's personal SDLC toolbelt for AI coding agents — from PRD to ship. Bundles the tracer-bullet workflow alongside TDD, code review, audits, and shipping skills.
AI-powered development tools for code review, research, design, and workflow automation.