start

Initialize project context and classify development tasks for coding sessions.

Updated Apr 13, 2026
One-click install
npx skills add https://github.com/MaplumeX/deepresearch --skill start-maplumex
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: start
Source: https://github.com/MaplumeX/deepresearch/tree/main/.agents/skills/start
Command: npx skills add https://github.com/MaplumeX/deepresearch --skill start-maplumex

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill removes the friction of starting or resuming development by automatically gathering project workflow, developer identity, git status, active tasks, and guidelines so the agent and developer share the same context before any changes are made.

Core Features & Use Cases

  • Session Initialization: Reads workflow guides and indexes, runs repository context scripts, and reports current task status to the developer.
  • Task Classification & Routing: Classifies incoming work as question, trivial fix, simple task, or complex task and routes it to direct edit, task workflow, or brainstorm as appropriate.
  • Task Workflow Orchestration: Creates task directories, writes PRD, performs focused research, injects code-spec context, activates tasks, implements changes, and runs quality checks.
  • Use Case: Use when starting a new coding session, resuming interrupted work, or when a user provides a new task description that must be classified and prepared for implementation.

Quick Start

Initiate a development session to load project workflow, current task state, and classify the next action for implementation.

Frequently Asked Questions about start

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I initialize project context and classify development tasks for an AI coding session?

To initialize project context, the system reads .trellis workflow files, spec indexes, and git status, then classifies incoming development tasks to route work to brainstorm, direct edit, or task workflows with context injection.

What is the best way to resume interrupted development tasks in a repository?

Resuming interrupted development tasks involves reading the current task state from task directories, restoring developer identity, and re-injecting code-spec context to seamlessly continue the task workflow orchestration.

How does task classification and routing work for incoming development work?

Task classification evaluates incoming work as a question, trivial fix, simple task, or complex task, then automatically routes it to direct edit, task workflow, or brainstorm workflows with appropriate context injection.

Do I need .trellis workflow files to start a development session?

Yes, .trellis workflow files are required because the session initialization reads these workflow guides and spec indexes to gather project guidelines and establish shared context before any code changes are made.

When should I use task workflow orchestration instead of direct edits?

Task workflow orchestration is used for simple or complex tasks that require creating task directories, writing PRDs, and performing focused research, whereas direct edits handle trivial fixes and questions.

Why does my AI coding session need context injection before implementing changes?

Context injection is needed to ensure the agent and developer share the same project workflow, developer identity, and active task guidelines before any changes are implemented or quality checks are run.