Chapter
Sources
- Anthropic — Prompting best practices, Prompting Claude Opus 4.8, Introducing Claude Opus 4.8; Best practices for Claude Code; Create custom subagents & Skills docs; Automate actions with hooks; Effective context engineering for AI agents; Effective harnesses for long-running agents; Writing effective tools for AI agents; Claude Code GitHub Action
- Multi-model — Anthropic The advisor strategy (+2.7pt / −11.9% cost) & Claude Code model-config/advisor docs;
opusplanwrite-ups; Aider architect/editor benchmark; cross-lab review reports (charlesjones.dev, dev.to, Milvus model-debate); MoA (Wang et al. 2406.04692) vs Self-MoA (Li et al. 2502.00674); Sakana AB-MCTS; GitHub Agent HQ;johannesjo/parallel-code - OpenRouter routing — OpenRouter Claude Code integration + Anthropic Skin docs/blog; Morph, TokenMix, mykolaaleksandrov.dev setup guides;
musistudio/claude-code-router - Agents vs one-shot builders — Claude Code vs Lovable/Bolt/v0/Replit/Base44 comparisons (lowcode.agency, builder.io, buildthisnow, nxcode, 2026)
- Playwright MCP (
@playwright/mcp, Microsoft) — official docs; Builder.io & qaskills setup guides - BDD/specs — AutomationPanda gherkin-guidelines-for-ai; ATDD with AI; Spec-Driven Development (Fowler; arXiv 2602.00180; BCMS 2026 guide)
- Vertical slicing — CODE Magazine AI-Assisted Greenfield Dev: Vertical Slices
- Terminal & session ops — Warp Agents docs (vertical tabs, notifications, code review, Oz orchestration); Run many Claude Code sessions in parallel (Towards Data Science); Claude Code-on-VPS + tmux guides (claudefa.st, virtua.cloud, codeongrass, Medium); Claude Code remote-session feature request (#13613)
- Agent Execution Layer — Northflank & Modal enterprise-deployment + sandbox guides; E2B/Modal/Daytona sandbox comparison (AgentMarketCap); Linear Agents + changelog (cloud coding sessions); Cyrus + Hookdeck (Claude Code as a Linear agent); daily.dev Huginn Linear coding agent; Claude Code GitHub Actions + GitHub Agent HQ;
ai-boost/awesome-harness-engineering - Effort /
ultracode/ dynamic workflows — Claude Code model-config docs + 2026 guides (claudefa.st, stacknotice) - DataCamp Claude Code Best Practices; rosmur 12-source synthesis; stevekinney AI-dev course; Addy Osmani Agent Harness Engineering; Viv Trivedy Anatomy of an Agent Harness
- Methodologies to study: Superpowers, Spec Kit, BMAD-METHOD, OpenSpec · Directories:
awesome-claude-code, Playwright MCP