What problem does it solve? Starting or resuming work in a Trellis-managed project requires manually reconstructing context: who you are, what git state you're in, which task is active, and which workflow rules apply. This Skill automates that session bootstrap so an AI agent begins every coding session with the correct project context and routing decision. ## Core Features & Use Cases - Session Context Loading: Runs .trellis/scripts/get_context.py to retrieve developer identity, git status, current task, active tasks, and journal location. - Workflow & Guideline Discovery: Loads the compact Phase Index, request triage rules, and package/spec guideline indexes from .trellis/spec/ so coding standards are known before editing. - Task Routing: Classifies the incoming request and routes to the right follow-up skill, such as trellis-continue for active tasks, trellis-brainstorm for new features, or trellis-before-dev before writing code. - Use Case: You open a repository mid-sprint and ask the AI to continue work. The Skill detects the active task, resumes it via trellis-continue recovery rules, and surfaces any pending Trellis update notice. ## Quick Start Ask the AI to start a new Trellis session and load the current project context before beginning any coding task.