kk — Claude Code Plugin
A development workflow plugin for [Claude Code](https://docs.anthropic.com/en/docs/claude-code) that gives your AI assistant a structured pipeline — from idea through design, implementation, code review, testing, to documentation. Part of the [claude-toolbox](https://github.com/serpro69/claude-toolbox) project.
Overview
kk — Claude Code Plugin
A development workflow plugin for Claude Code that gives your AI assistant a structured pipeline — from idea through design, implementation, code review, testing, to documentation. Part of the claude-toolbox project.
Why kk? I have jj and kk mappings in nvim to go back to normal mode, and it seemed like a low-conflict/easy-to-type option for a plugin-prefixed skill names. We can also pretend it's an acronym for "klaude kode", if you need a better reason for the plugin naming.
Installation
The plugin is installed automatically when using the claude-toolbox template. To install manually:
/plugin install kk@claude-toolbox
All skills appear as /skill-name in the slash command menu (annotated with (kk)). No additional configuration needed.
Skills
| Skill | What it does |
|---|---|
| design | Turns an idea into design docs, an implementation plan, and a task list in docs/wip/. Asks refinement questions, then documents everything a developer needs to start coding. |
| implement | Executes a task list from docs/wip/ with batched steps and code review checkpoints between batches. Updates task status as it goes. |
| test | Generates tests following project conventions: table-driven, integration, mocking, property-based. Runs the full suite and reports coverage. |
| document | Updates ARCHITECTURE.md, TESTING.md, and records ADRs for non-obvious decisions made during implementation. |
| dependency-handling | Fires before calling a library/SDK/API or adding a dependency. Forces a capy/context7 lookup instead of guessing signatures or behavior. |
| review-code | Reviews git changes for SOLID violations, security risks, and code quality. Language-specific checklists for Go, Java, JS/TS, Kotlin, and Python. Standard and isolated modes. |
| review-spec | Compares implemented code against design/implementation docs. Finds spec deviations, missing implementations, and outdated docs — in both directions. |
| review-design | Pre-implementation review gate. Evaluates design docs for completeness, internal consistency, and technical soundness before code is written. |
| merge-docs | Merges two competing design docs for the same feature into one unified document, resolving conflicts and preserving the best of both. |
| chain-of-verification | Makes Claude fact-check its own answers. Standard mode (prompt-based) or isolated mode (independent sub-agents). For high-stakes accuracy. |
Development Workflow
The skills are designed to work together in a pipeline:
design → review-design → implement → review-code → test → document
- design — design docs + implementation plan + task list
- review-design — evaluate design docs for completeness and technical soundness before writing code
- implement — execute tasks with review checkpoints
- review-code — review code for SOLID violations, security risks, and quality issues
- test — verify and validate
- document — update docs
During implementation, dependency-handling is pulled in whenever you touch an external library, SDK, or API — it routes you through capy/context7 instead of letting you guess. review-spec verifies code matches design/spec and detects deviations — use during or after implementation.
Utilities: merge-docs reconciles competing design docs into one unified document. chain-of-verification adds self-verification for high-stakes accuracy at any stage.
Commands
| Command | Invocation | Description |
|---|---|---|
| Code Review (isolated) | /kk:review-code:isolated | SOLID code review with independent sub-agents |
| CoVe (standard) | /kk:chain-of-verification:default [question] | Chain-of-Verification with prompt-based isolation |
| CoVe (isolated) | /kk:chain-of-verification:isolated [--explore] [--haiku] [question] | CoVe with true sub-agent isolation |
| Implementation Review | /kk:review-spec:default [feature] | Verify code matches design/implementation docs |
| Implementation Review (isolated) | /kk:review-spec:isolated [feature] | Spec conformance review with independent sub-agent |
| Design Review | /kk:review-design [feature] [scope] | Review design docs for quality and technical soundness |
| Design Review (isolated) | /kk:review-design:isolated [feature] [scope] | Design review with independent sub-agents |
| Migrate from Task Master | /kk:migrate-from-taskmaster:migrate | One-time migration from Task Master MCP |
| Sync Workflow | /kk:sync-workflow:sync-workflow [version] | Update template-sync from upstream |
Hooks
| Hook | Trigger | Description |
|---|---|---|
| Bash validation | PreToolUse on Bash | Blocks commands touching .env, .git/, node_modules, build/, dist/, venv/, and other sensitive paths |
Profiles
The plugin ships per-domain profiles under profiles/ (e.g., go, k8s, python). Profiles provide language- and framework-specific content to every workflow skill — review checklists, implementation gotchas, design prompts, test validators, and doc rubrics.
Some profiles vendor content from external upstream repositories. The Go profile, for example, vendors from samber/cc-skills-golang via a manifest-driven pipeline (make vendor-go). See the Vendored profile content section of CLAUDE.md for the workflow, and the Profile Conventions section for the full authoring contract.
Upgrading from Template-Managed Skills
If you're upgrading from a version before the plugin system (< v0.5.0):
- Skills remain unprefixed:
/design(annotated with(kk)in the menu) - Commands are now namespaced:
/project:chain-of-verification→/kk:chain-of-verification:default - The template-sync workflow handles migration automatically on next sync
- After merging the sync PR, run
/plugin install kk@claude-toolbox