ai-playbook

muriloscigliano/ai-playbook
★ 0 stars JavaScript AI/LLM Updated 6d ago
78 engineering patterns + 17 design principles + structured data layer for building AI-first products. MCP server, CLI, web UI.
View on GitHub → 🔍 Audit Wallet Slippage →

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "ai-playbook": {
      "command": "npx",
      "args": [
        "-y",
        "muriloscigliano/ai-playbook"
      ]
    }
  }
}

README Excerpt

> **Production-hardened agent patterns, reverse-engineered from real systems** — not another ReAct-and-RAG explainer. > The core (63–78) is the stuff that only shows up once agents hit production: deferred tool loading, hierarchical memory files, a skills system, the one-core-many-surfaces bridge pattern, and tool-result tombstones. Wrapped around it: 78 engineering patterns, 17 design principles, 9 UX patterns, and a typed data layer that connects them.

Tools (20)

aiTasksallRelationsautonomyLevelscapabilitiesdiagnose_agentdiagnose_uxget_build_guideget_capabilityget_design_principleget_design_sectionget_patternget_ux_patternhumanTaskslist_capabilitieslist_patternsproblemDiagnosesprojectBlueprintsrecommend_capabilityrecommend_designrecommend_patterns

Topics

agentic-aiaiai-agentsdesign-patternsdesign-principlesllmmcpmulti-agentplaybookproduction-readyragux-patterns