Uses power tools
Uses Bash, Write, or Edit tools
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.
|English|日本語|
deckrd is a document-driven workflow for progressively documenting and organizing requirements through implementation decisions.
It manages the following documents as separate, distinct layers:
Prerequisites: AI execution environment (Claude Code, etc.) and access to a compatible LLM
deckrd consists of three integrated components that cover the full development lifecycle:
| Component | Command | Role |
|---|---|---|
| deckrd | /deckrd <command> | Planning: Goals → Tasks (design phase) |
| deckrd-coder | /deckrd-coder <task-id> | BDD-style implementation (coding phase) |
| IDD Framework | /idd/issue:new, /idd-pr | GitHub Issues / PRs (execution phase) |
Phase 1: Issue Definition (IDD)
/idd/issue:new → GitHub Issue → Branch
Phase 2: Planning (deckrd)
/deckrd init → module → req → [dr] → spec → impl → tasks
Phase 3: Implementation & Release (deckrd-coder + IDD)
/deckrd-coder T01 → BDD impl → /idd-commit-message → /idd-pr → PR
# Add marketplace if needed
claude plugin marketplace add aglabo/deckrd
# Install deckrd
claude plugin install deckrd@deckrd
deckrd directory to your Agent Skills directory# Phase 1: Create Issue (IDD)
/idd/issue:new
# Phase 2: Plan with deckrd
/deckrd init myProject shell
/deckrd module myProject/feature
/deckrd req
/deckrd spec
/deckrd impl
/deckrd tasks
# Phase 3: Implement and release (deckrd-coder + IDD)
/deckrd-coder T01-01 # Implement task T01-01
/deckrd-coder T01-02 # Implement task T01-02
# ... (run for each task)
/idd-commit-message
git commit -m "..."
/idd-pr
| Command | Description |
|---|---|
init <project> <project-type> | Bootstrap project (run once per project) |
module <namespace>/<module> | Create module directory and set as active module |
req | Create requirements definition |
spec | Create specification document |
impl | Create implementation criteria |
tasks | Create implementation task list |
dr --add | Record Decision Records (optional) |
status | Check workflow progress |
For details, see:
The implementation layer is not a place to write actual code. It records the following aspects of implementation decisions:
deckrd does not aim to:
deckrd is purely an assistive framework for organizing thoughts and decisions.
deckrd-coder is an optional plugin for deckrd users.
It takes tasks.md generated by deckrd and automatically implements them using a strict BDD (Behavior-Driven Development) process.
Completely optional: You can use deckrd alone to generate tasks and implement them manually at your own pace.
deckrd-coder is available from the deckrd marketplace:
claude plugin install deckrd-coder@deckrd
npx claudepluginhub aglabo/deckrd --plugin deckrdDeckrd task implementation skill - automatically extracts specified tasks from tasks.md and implements them using strict BDD process. Internally executes bdd-coder agent. Optimizes token usage with project memory & serena-mcp.
IDD (Issue-Driven Development) and SDD (Spec-Driven Development) framework with custom slash commands and agents for systematic AI-assisted development workflows
Design doc driven task management for human-agent collaboration
Ultra-compressed communication mode. Cuts ~75% of tokens while keeping full technical accuracy by speaking like a caveman.
Comprehensive UI/UX design plugin for mobile (iOS, Android, React Native) and web applications with design systems, accessibility, and modern patterns
Memory compression system for Claude Code - persist context across sessions
Standalone image generation plugin using Nano Banana MCP server. Generates and edits images, icons, diagrams, patterns, and visual assets via Gemini image models. No Gemini CLI dependency required.
Multi-model consensus engine integrating OpenAI Codex CLI, Gemini CLI, and Claude CLI for collaborative code review and problem-solving.