
Watch How an AI Agent Thinks
在可交互的时间轴上探索三个 AI 智能体的运行过程,展示规划、工具调用和故障恢复。
douvy · HN
Educational tool for understanding AI agents. Code: https://github.com/douvy/how-agents-think
完整作品展
技术栈
60 projects

在可交互的时间轴上探索三个 AI 智能体的运行过程,展示规划、工具调用和故障恢复。
douvy · HN
Educational tool for understanding AI agents. Code: https://github.com/douvy/how-agents-think

比较 git、Jujutsu 和 GitButler 在 Claude Code 和 Codex 代理上的性能。
videlov · HN
I was interested in answering this question so I built a benchmark comparing git, jj and gitbutler in agentic context https://vcbench.dev/ Disclaimer - I am a co-founder of GitButler

Publish HTML and Markdown artifacts from any AI agent with version history and team access.
@slashdisplay · X
One feature, two agents, one URL. Carl published v1 from Codex. Claude Code iterated to the version that shipped – every hand-off attributed in the audit log. A 3-minute walkthrough of a real workflow:

在Google日历中调度AI代理任务
matt413 · HN
Schedule tasks for your AI agents from Google Calendar

实时仪表板展示 AI-agent 公司的 ARR、MRR、任务和活动。
pro_methe5 · HN
I scrape an $8.5M-ARR company run by AI agents and chart it live

ForgeLab Brain:多个AI智能体协作编程,自动验证和修复错误
@ForgeLab_Brain · X
Open beta launched. See more here: 🌐 #VibeCoding #BuildInPublic #IndieHacker #AIdev

粘贴两个 AI agent 版本,对比治理权限的变化。
stoicstoic · HN
Show HN: Diff two versions of an AI agent and catch silent permission changes

按职业场景语义搜索 AI Skills,保存收藏集,一键装入 Claude Code。
SkillForge — Claude Skill 发现与分发平台,按职业场景组织 5700+ skill 覆盖 30 个垂直领域,一行命令装到 Claude Code / Cursor,登录后可留存自己的工具集

在3D中探索AI代理编排,实时监控项目健康和日志。
@patologico · X

Tradovate AI 风险管理,根据您的风险规则自动阻止订单。
@SanuCodes · X
still building-

在多个运行时中设计、验证和比较AI代理部署,具有内置治理功能。
@paulrodturner · X

用于在 git worktrees 中并行运行多个 AI coding agents 的桌面应用
vuphanse