Dearest

Production-grade Agent development methodology derived from Claude Code. 11-dimension framework covering tool design, system prompts, permissions, multi-agent orchestration, token economy, memory, extensibility, conversation loop, workflow modes, task lifecycle, and observability. Supports architecture design, implementation guidance, and agent review. Trigger on "Agent design", "build an agent", "AI agent", "tool design", "system prompt architecture", "agent review", "multi-agent", "compaction", "agent workflow modes", "background tasks", "agent observability", or any agent development concern.

Use when starting debug tasks, fixing bugs, building new features, performance optimization, dependency migration, or automated testing. Also use after completing significant problem-solving sessions involving undocumented behavior, repeated failed attempts, or multi-step debugging.

写"橙皮书"风格的技术书/实战手册——从选题、章节大纲、分章写作、审校到渲染成出版级 A4 PDF。当用户要把一个技术主题（某个工具、框架、概念）做成一本结构化、可下载、读起来不像 AI slop 的免费技术书/指南/手册/电子书时使用。触发词：橙皮书、技术书、实战手册、电子书、PDF 教程、把 X 做成一本书。

Runs a local daemon connecting your machine to the Claude Code in Slack server. Receives tasks from Slack, drives Claude Code locally via @anthropic-ai/claude-agent-sdk, and streams results back. Use to set up credentials, start/stop the daemon, and check status. use when "claude-code-in-slack", "setup slack bridge","connect to slack", or any mention of managing the Claude Code in Slack daemon.