rawr-ai

Mental model for a sealed service whose context and authorship narrow toward operation handlers.

Mental model for locating work in RAWR's blueprint kinds while keeping semantic direction, ownership, and structural proof distinct.

Mental model for Nx as the repository graph and scheduler while project owners retain their qualified build and verification work.

Mental model for one conventional installed Oclif application that owns CLI identity while commands retain qualified capability boundaries.

Mental model for a first-party Oclif command plugin as one host-composed terminal projection on public package boundaries.

Mental model for an application as the concrete selection and realization boundary of the product chain.

Mental model for inert reusable support that carries no product, runtime, or projection authority.

Mental model for projecting resolved Habitat applications into the Nx scheduler without acquiring policy authority.

This skill should be used when the user asks about "target architecture", "greenfield spec", "pure SPEC", "decision packet", "migration slice", "architecture decision", "spike to spec", "current vs target", "hybrid to target", "indirection audit", "dependency ordering", "cutover validation", "no legacy left", "legacy cutover", or needs guidance on defining target architecture, creating specs from exploration, resolving architectural ambiguity, planning migrations from current to target state, or validating cutover plans.

This skill should be used when the user asks about "dev-auto", "autonomous development", "orchestrator", "worktree lifecycle", "structured output", or when running dev-auto-* commands or any autonomous loop that relies on orchestrator-owned worktrees, explicit inputs, and structured JSON outputs. Auto-safe development contract for orchestrated dev/review/fix workflows.

This skill should be used when the user asks about "git worktree", "parallel checkout", "worktree cleanup", "create worktree", "remove worktree", "worktree path", or when a workflow involves git worktrees, parallel development, multiple checkouts, or edits that must happen in the intended worktree path. Reusable workflow for safely creating, entering, and working within git worktrees (plus clean removal).

This skill should be used when the user asks about "Linear method", "Linear issue", "milestone sizing", "triage", "Linear project", "acceptance criteria", "issue scope", or needs guidance on shaping high-quality Linear work (issues, milestones, projects) as clear, right-sized units of value and behavior. Covers planning, writing, and reviewing Linear items around scope, sizing, triage, acceptance criteria, and human-first clarity. Includes references and templates.

This skill should be used when the user asks about "parallel development", "multiple agents", "worktree + Graphite", "parallel issues", "agent isolation", or needs to coordinate multiple agents or issues in parallel using git worktrees + Graphite stacks (gt). Each agent isolates work in its own worktree + local stack and avoids global restacks. Canonical, reusable parallel development workflow.