From sage
Start here. Sage reads project state, routes via keywords, classifies intent, and guides you to the right workflow.
How this skill is triggered — by the user, by Claude, or both
Slash command
/sage:sageThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
RULES (apply to every step — non-negotiable):
RULES (apply to every step — non-negotiable):
Sage's intelligent entry point. Assess the project and guide the user.
Scan .sage/work/ for active initiatives (read frontmatter: title,
status, phase). Scan .sage/docs/ for project-level artifacts.
Read .sage/decisions.md for recent context.
Present what you found, then structured options based on context.
If work is in progress:
Sage: [Project name] — [feature] is in progress, [phase] phase.
[1] Continue [feature] — resume from [next step] [2] Start something new [3] Review what's been done
If no work in progress but artifacts exist:
Sage: [Project name] — no active work. Previous: [list initiatives].
[1] Start a new task — describe what you want to build [2] Review existing artifacts [3] Learn the codebase
If fresh project:
Sage: Fresh project, no work in progress.
[1] Build something — describe what you want to create [2] Learn the codebase first [3] Something else — describe what you need
Based on user's choice or free-form input, classify scope and route:
For complex routing or gap detection, read the sage-navigator at
sage/core/capabilities/orchestration/sage-navigator/SKILL.md.
$ARGUMENTS
npx claudepluginhub xoai/sage --plugin sageSage process framework — constitution, routing, interaction zones, enforcement rules. Auto-loads to provide process enforcement, keyword routing, and structured interaction patterns for all Sage workflows. This is the always-on layer that ensures quality even when specific workflow skills are not loaded.
Analyzes project state to suggest next actions based on pipeline status, artifacts, and context. Helps developers decide what to do next.
Auto-analyzes project state including tasks, source code, git conflicts, and agents to recommend 1-2 optimal skills. Use on 'what's next?' questions or /workflow trigger.