- 发布时间
过去我们手动拆任务、一个个启动 agent。如今 Claude Code 的 dynamic workflows（ultracode 模式）会自己写出编排脚本、调度一整支 subagent 队伍——本文通过构建 iread RSS 阅读器的三个真实工作流来展示，并把 agent 之间的异步通信留作下一个待解问题。
Multi-agent
全部文章
- workflow (12)
- linux (11)
- ai (11)
- tools (10)
- arch-linux (8)
- productivity (8)
- opencode (7)
- ide (7)
- claude-code (7)
- ai-coding (7)
- multi-agent (7)
- wayland (5)
- neovim (5)
- window-manager (4)
- tiling (4)
- coding (4)
- subagents (4)
- sway (3)
- vibe-coding (3)
- mcp (3)
- cli (3)
- journey (3)
- vim (3)
- configuration (3)
- development (3)
- ai-agent (3)
- rust (3)
- i3 (2)
- river (2)
- plugins (2)
- skills (2)
- ikanban (2)
- vscode (2)
- editor (2)
- dotfiles (2)
- setup (2)
- guide (2)
- introduction (2)
- reflection (2)
- chat (2)
- automation (2)
- misc (2)
- hyprland (2)
- browser (2)
- surfingkeys (2)
- blog (2)
- codex (2)
- budget (2)
- video-generation (2)
- development-tools (2)
- deepseek (2)
- vibe-kanban (2)
- parallel-workflow (2)
- task-scheduling (2)
- qwen (2)
- 2025 (1)
- niri (1)
- scrollable-tiling (1)
- screen-sharing (1)
- webrtc (1)
- desktop-environment (1)
- x11 (1)
- keyboard-first (1)
- jetbrains (1)
- copilot (1)
- autohotkey (1)
- windows (1)
- experience (1)
- daily-use (1)
- agent (1)
- security (1)
- openness (1)
- life (1)
- summary (1)
- philosophy (1)
- claude (1)
- specialized-agents (1)
- development-workflow (1)
- github (1)
- godot (1)
- game (1)
- dev (1)
- laptop (1)
- i3wm (1)
- nvidia (1)
- wlroots (1)
- nixos (1)
- declarative (1)
- reproducible (1)
- deploy (1)
- oss (1)
- bot (1)
- js (1)
- typst (1)
- typesetting (1)
- document (1)
- template (1)
- academic (1)
- keymaps (1)
- chrome-devtools-mcp (1)
- verification (1)
- token (1)
- pricing (1)
- openrouter (1)
- tts (1)
- python (1)
- manim (1)
- cache-hit (1)
- tmux (1)
- academic-research (1)
- dynamic-workflows (1)
- ultracode (1)
- orchestrator (1)
- superpowers (1)
- danmu (1)
- streaming (1)
- real-time (1)
- bilibili (1)
- md2video (1)
- agent-first (1)
- ai-workflow (1)
- markdown (1)
- cui (1)
- gesture-recognition (1)
- magic-mouse (1)
- multi-touch (1)
- multi-provider (1)
- kimi (1)
- glm (1)
- keyboard-navigation (1)
- gui (1)
- evdev (1)
- 发布时间
在 OpenCode 之上构建工具和工作流几个月之后，我们把日常使用切回了 Claude Code。原因是具体的：GPT 调整账号安全策略之后模型性价比发生变化、把 Claude 走 OpenCode 带来的封号风险、prefix cache 失配带来的周配额损耗，以及一个并没有带来质变的 browser 包装实验。- 发布时间
vibe coding 的下一步，不是彻底移除人类，而是尽量把人从每个内部开发循环里移出去。当 agent 能自己完成更多验证时，多 agent 系统就不会继续被同一个人卡住。- 发布时间
对 AI agent token 获取的两条主要路径进行实际分析——按量付费的 API 方式与订阅制的代码计划——附真实成本数据，并根据单 agent 还是多 agent 场景给出具体建议。- 发布时间
面向刚接触 agent coding 的读者，介绍我们最新的四层多 agent 工作流：更便宜的模型访问、OpenCode runtime、Superpowers 自动化，以及 iKanban 审查界面。