Create GitHub pull requests — picks the most token-efficient available method (gh CLI, GitHub MCP, or git push), and sources the PR body from an existing spec/plan instead of re-analyzing the diff
Bootstrap any project's AI workspace — detect the current tool, extract hardcoded paths from skill/command/agent files, write a central workspace.md, update all files to use config keys
Map the whole system into a `docs_context` documentation layer from a full codebase scan — creates or reconciles the business and system docs. Supports single-service repos and multi-service monorepos with orchestrator node.
Feature & chore planning — research codebase, design options, write a structured spec/implementation plan to the project specs directory
Implement an approved spec — reads plan from specs_dir, executes phase by phase, updates checkboxes, verifies after each phase, and commits completed work
Identify, choose, apply, and audit design patterns — names the patterns a codebase already uses, evaluates fit before recommending new ones, applies them incrementally during implementation, and reviews code or a design doc for opportunities and anti-patterns
Prime — load project context by reading docs, the docs_context layer, core docs, and TODO before starting any task
Define acceptance test cases in DSL format at confirmed seams — happy paths, edge cases, error scenarios, and authorization, sequenced into a build order