agentic▌

AI-driven engineering workflows with eval-first execution, task decomposition, and cost-aware model routing. \n \n Defines an eval-first loop: establish baseline evals before implementation, then re-run post-execution to measure deltas and catch regressions \n Decomposes work into 15-minute units with single dominant risks, independent verifiability, and clear done conditions \n Routes tasks by complexity: Haiku for classification and boilerplate, Sonnet for implementation, Opus for architecture

The paradigm shift from screen-centric to relationship-centric design.

Optimized patterns for AI agent workflows, Git integration, MCP servers, and multi-agent orchestration. \n \n Covers essential commands across Claude, Gemini, and Codex CLIs, plus keyboard shortcuts for task cancellation, history search, and plan mode toggling \n Includes session management techniques (resume, rename, aliases) and Git workflows for auto-generating commits, PRs, and managing parallel branches with worktrees \n Provides MCP server configuration guidance (Playwright, Supabase, Fire