Task implementation workflow for the /m:dev command. Provides two execution modes: single-task mode (7-step workflow for one task at a time) and multi-task mode (wave-based orchestration with parallel subagents for feature groups, ranges, and use cases).
From mnpx claudepluginhub molcajeteai/plugin --plugin mThis skill uses the workspace's default tool permissions.
references/multi-task-mode.mdreferences/single-task-mode.mdSearches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Searches prompts.chat for AI prompt templates by keyword or category, retrieves by ID with variable handling, and improves prompts via AI. Use for discovering or enhancing prompts.
Analyzes unfamiliar codebases to generate structured onboarding guides with architecture maps, key entry points, conventions, and starter CLAUDE.md.
Execution modes for implementing tasks from a feature's tasks.md file.
| Mode | Trigger | Description |
|---|---|---|
| Single-task | 1 pending leaf task selected | Sequential 7-step workflow: gather context, clarify, plan, code, test, run tests, quality gate |
| Multi-task | 2+ pending leaf tasks selected | Wave-based orchestration: plan all tasks upfront, execute in dependency-ordered waves with parallel subagents |
| File | Purpose |
|---|---|
| references/single-task-mode.md | Full single-task workflow (Steps 1-7, quality gate with 4 parallel agents) |
| references/multi-task-mode.md | Multi-task orchestration (wave construction, parallel execution, session state) |