An automated multi-agent coding workflow that orchestrates Software Design Document (SDD) creation, Multica issue generation, dependency-aware task dispatching, and strict completion logging.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install multica-sdd-workflow
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install multica-sdd-workflow using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
The Multica SDD Workflow is an advanced automation process designed to orchestrate structured multi-agent development pipelines. By integrating Software Design Documents (SDD) with the Multica task manager, this setup stands out among Openclaw Skills as a complete solution for robust, quality-driven codebase scaling. It bridges high-level architecture with programmatic task execution, guaranteeing strict compliance with project design specifications before any code is merged.
This workflow is built for engineering teams and AI agents that require strict API-first designs, detailed spec documentation, and test-driven cycles. It coordinates squads of specialized coding agents, tracking execution progress dynamically and ensuring repository files like CLAUDE.md and tasks.md reflect the living state of the software at all times.
.multica/squad.md to establish online agent availability.To initialize this workflow within your workspace, configure your environment with the following files and directories:
# Verify agent status and squad availability
multica agent list --output json
multica squad list --output json
# Create a new project in Multica
multica project create --name "Your-Project-Name" --output json
# Create a development squad and add members
multica squad create --name "Squad-Name" --description "Task force" --leader "$LEADER_ID" --output json
multica squad member add "$SQUAD_ID" --member-id "$EXECUTOR_ID" --role member
.multica/squad.md with the schema detailed in the data schema section below.specs/
└── <NNN>-<feature-name>/ # NNN is a 3-digit serial number, e.g., 001-auth
├── spec.md # Feature specification requirements
├── plan.md # Architectural and technical implementation plan
└── tasks.md # Granular task list checkpoints
| Field | Value Description |
|---|---|
| project-id | Unique UUID for the project in Multica |
| squad-id | Unique UUID for the agent squad in Multica |
| leader-id | UUID of the coordinating leader agent |
| executor-id | UUID of the executing agent |
| executor-name | Readable string name of the executor agent |
| work-dir | Absolute path to the repository directory |
| test-cmd | CLI test invocation string (e.g., npm test) |
| typecheck-cmd | CLI static-analysis command (e.g., npm run typecheck) |
Updates are written directly to CLAUDE.md to track high-level roadmap status, listing current percentage metrics and detailed task tables of complete and pending items.
Loading
An AI-powered automation skill that transforms raw product images into fully optimized, high-converting mobile H5 detail pages in under 10 minutes.

An AI-powered intellectual assistant that navigates Bertrand Russell's 'A History of Western Philosophy' to connect historic thought with modern context.

An interactive AI agent skill based on Aldo Leopold's classic book, guiding users through modern conservation, phenology, and the famous Land Ethic.

A shared design system that provides over 30 token CSS variables, multiple layout templates, and a visual showcase to ensure UI consistency across automated assets.

An Openclaw Skills configuration to generate, refine, and localise clean, family-friendly dad jokes, puns, and themed humor sets.

An AI-powered cognitive science agent skill that models the five evolutionary breakthroughs of the brain to bridge the gap between neuroscience and modern artificial intelligence.








































