Robocoppo

interested in trading, consumer finance, subprime lending and fintech

[宝玉 @dotey]

很多人不知道该怎么用好 Agent 的 /goal 功能，也就是说给 Agent 一个目标，让它长时间运行，直到目标完成为止。

其实没你想的那么复杂，注意几个点：
1. 你的目标是什么
2. 如何验证结果
3. 停止条件

比如说我这两天做的一个性能优化的任务，Fable 5 帮我把视频转录性能优化了2倍多（图2），提示词很简单（图1）：
> /goal 帮我优化当前 cli 的转录大视频的性能，在遇到像这样大体积的视频时，需要优化转录性能，请以 Moss 模型测试这个视频（英文为主，多语言）转录，建立基准，然后分析性能瓶颈，尝试优化，直到你觉得已经没有优化空间了。注意你的主要任务是分析、编排和验证，具体任务尽可能交给 subagent（Opus5）去执行

首先用 /goal 表示这是一个需要长时间执行的任务，需要反复执行，不能运行一会就结束了。

然后给它一个视频让它先自己跑一遍转录，记录一下关键数据，建立基准。

基于转录时收集的数据，Agent 自己可以去分析原因，去自己优化，优化完成后再去跑一遍，记录数据，对照前面的基准看是更好了还是更坏了。

结束条件是它自己觉得已经没有优化空间了就结束。之所以我没给它一个具体指标，是因为我也不知道能优化多少，如果指标太容易达到，它一轮可能就结束了；如果指标太难超出物理极限也没意义，反而可能会出现为了优化去做一些极端的事情。

最后一句让它开subagent执行子任务是因为 Fable 5 太贵，全程 Fable 5 用不了多久就要额度不够了，加了这句就耐用多了，而且质量也挺好。

---

还有些时候，想到一个新的技术方案，但并不知道这方案是不是有效，那也可以让它开个worktree，去验证一下是不是靠谱，看数据是更好还是更坏，如果没提升就没必要做了。（参考图3）

[Gloria @changloria0816]

https://t.co/EOgpPO1YAu

[Roan @RohOnChain]

Don't waste 2 years learning to become an AI agentic engineer in 2026.

Andrew Ng, the godfather of AI, gave the complete playbook to become one from scratch.

1 hour course. Free:

• 00:00 - AI agent basics
• 12:12 - AI Agentic workflows & design patterns
• 53:27 - Practical tips for building AI agents
• 1:20:30 - self-improving AI agent loops
• 1:30:19 - multi-agent AI systems

I watched it last night.

Halfway through, I realized I could get into Anthropic in weeks, not years.

Bookmark now. Watch it. Then build your own AI agent with the guide below.

[雪踏乌云 @Pluvio9yte]

今天阅读了两篇关于harness的文章，算是比较基础的，也很有启发。

关于harness一个比较基础的介绍，很清晰易懂，从model出发，讲解由model到agent的过程中，需要哪些比较基本的harness功能
https://t.co/SQBKe08hlc

openai的外部harness实践，如何从0开始用ai写代码，核心思想就是“创建一个让 AI 能持续正确工作的环境”
当然其中的一些思想现在已经被codex的harness做好了，但是仍然有很多经典的可以阅读的地方。
https://t.co/WB7NwBu9Ac

[Lunar @LunarResearcher]

Google just released a free 2-hour+ course on full agent engineering.

How to go from one prompt to a system that runs while you sleep:

38:46 - Build your first AI agent
54:46 - Connect MCP tools
1:12:43 - Run agents with four loop patterns
1:20:57 - Turn agent loops into graphs
2:22:31 - Build the complete autonomous system

Most people are still building one agent and stopping there.

Google is already teaching the full stack:

Agents → Tools → Loops → Graphs → Autonomous Systems

Single agents are the old workflow.

Systems that run without you are the new one.

This course is worth more than most paid agent engineering bootcamps.

Bookmark and watch it today

Then read the full graph engineering playbook below

[Anatoli Kopadze @AnatoliKopadze]

Head of Claude Code:

"85% of our engineers are running dozens or hundreds of agents. The way you do it is graph engineering."

In 40 minutes he explains how a single engineer now does the work of a whole team, how far this has gone inside Anthropic and where it goes next.

This is something you can't skip if you don't want to be left behind.

Watch it, then read the full guide on graph engineering below.

[Xudong Han @Xudong07452910]

这篇文章，我真心建议所有使用 Claude Code 或者 Codex 的人都看一遍。

我之前写过几次类似的感受：模型能力提升后，我们还在沿用旧模型时代的提示习惯，给 Agent 塞满规则、示例和各种「不要做什么」，很多时候反而会限制它。

Anthropic 最近把 Claude Code 面向 Opus 5 和 Fable 5 的系统提示词删掉了 80% 以上，代码评测几乎没有下降。

他们发现，过多的规则很容易互相冲突，Claude 还得先花力气判断该听哪一条。固定示例也会缩小模型的探索范围，本来能自己判断的事情，最后被框死在预设路径里。

现在的思路变得简单很多：

让模型结合代码库自己判断；把工具接口设计清楚；验证和代码审查需要时再通过 Skills 加载；CLAUDE.md 只保留仓库里真正特殊、很难从代码中直接发现的坑。

我觉得 Context Engineering 正在进入一个新阶段。

过去像是给 Agent 写一本厚厚的员工手册，现在更像是给它搭建一个清晰的工作环境。

模型越强，真正有价值的上下文可能越短：明确的边界、独特的经验、关键的参考，剩下的交给它自己判断。

很多过去有效的「最佳实践」，现在可能已经从护栏变成了脚镣。

[阿良｜AI 工作流 @RealYDT]

很多 Agent 做不长久，不是模型不够强，而是它根本记不住信息之间的关系。
IBM 最近放出一套 1 小时免费课，从零讲清 Agentic Knowledge Graph：
怎么搭第一张知识图谱、怎么给 Agent 加长期记忆，以及怎么用图结构协调多个 Agent。
重点不是再教你调用一个框架，而是让 Agent 从“搜到什么答什么”，变成能理解关系、保留上下文、持续积累知识。
别信“1 小时替代 10 门付费课”这种宣传，但这套内容确实能帮你快速判断：知识图谱是不是下一步值得补的 Agent 能力。
模型决定上限，记忆和知识结构决定它能不能真正工作。
https://t.co/Ic4tUME40w

[codila @0xCodila]

This free 6-hour course gets you Anthropic official "Claude Certified Architect" certificate

with it you'll be hired at $250K/year in Anthropic's ecosystem

10:17 - set up your first Claude SDK in 10 min
47:36 - agentic loop foundations
1:52:53 - graph engineering
2:50:00 - tool use - make your agent act, not just think
4:15:46 - advanced agent patterns

after watching the full freeCodeCamp tutorial (12 hours) - go to Anthropic's website and take the exam - you're ready

watch it today - then read how to become a knowledge graph architect in the article below ↓

[Robocoppo @robotcopo]

@TheEconomist The economist is a shame

[Ragnar @RoaringRagnar]

Here‘s the full Economist interview of Elon Musk (84 minutes).

[Robocoppo @robotcopo]

@TheEconomist @elonmusk @zannymb I like Elon, and I can't stand hypocritical media. Thanks, Elon, for speaking up for us!

[阿良｜AI 工作流 @RealYDT]

很多人一上来就让 Claude Code 干活，结果越用越乱。

真正应该先做的，不是写代码，而是把项目里的 ".claude/" 配好。

最基础的一套，至少包括：

→ "CLAUDE.md"：告诉它项目目标、技术栈和工作方式
→ "rules/"：把规则拆开管理，项目越大越有用
→ "commands/"：沉淀常用流程，别每次重复输入
→ "skills/" 和 "agents/"：把任务自动化，交给专门的子代理
→ "settings.json"：控制权限、行为和执行边界

".claude/" 不是一个普通文件夹。

它决定了 Claude Code 到底是在“帮你干活”，还是在“边干边猜”。

前期多花 10 分钟配置，后面能少踩一堆坑。

[Raytar @Raytar]

Don't waste 2 years learning to code.

The man who built Claude Code tells you what to learn instead.

82 minutes. Free:

24:03 - Anthropic's real engineering rules
27:44 - $100k+/month on Claude
40:42 - What to learn instead of coding

Writing code is the old job. Building loops is the new one.

Watch it. Then build your own loop with the guide below.

[rari @0xwhrrari]

Andrew Ng just released a 2-hour course

On building agentic skills from scratch with Anthropic:

00:00 - How to build agent skills with Claude
22:32 - Claude pre-built skills for AI agents
41:07 - Agentic skills vs tools, MCP, and subagents
01:06:06 - Skills for long-running agents

This 2-hour watch can replace 10 paid courses on building agents

Taught with Anthropic themselves

Bookmark and watch it tonight

Then read the article below

[Anatoli Kopadze @AnatoliKopadze]

Andrew Ng just dropped a 3-hour course on how to become an AI Engineer in 2026:

00:00 - build agentic AI systems from scratch
04:25 - where AI engineering is actually headed
23:38 - the full AI prompting course
2:52:17 - build a working app with AI in 30 minutes

This 3-hour watch replaces any $1000 course you could pay for.

Watch it today, then go deeper with the full guide on building AI agents below.

[李韭二 @li9292]

假如你还在问什么是loop Engineering.一张图看懂