情绪过山车

娴呭枩娣辩埍

[情绪过山车 @dpezzy903]

@0xForce_ he could be found court against today teammates, all easy like post shoes

[plusxbt @plusxbt]

I don't understand why everyone isn't doing this yet

Anthropic documented a Claude Code feature where the STRONGEST model only kicks in at key moments instead of running the whole time

The setup: Agent Teams in Claude Code

Rather than one session doing everything, the work splits across a team:

Lead (Opus 5.5, high effort): the main session, creates the tasks and runs the team

Teammates (Sonnet 5.5, medium effort): each owns its own files, UX or backend

Advisor (Fable 5.1): consulted before picking an approach, on a repeated error, before done

Shared task list: teammates claim tasks, file locks stop two of them grabbing the same one

Start the lead from the terminal:

claude --advisor fable

Then tell it: Spawn three teammates, one on UX, one on the backend, one playing devil's advocate. Use Sonnet for each teammate

Paste this to set up your workspace:

In settings.json, add CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1 to the env block, set effortLevel to high

Define roles in .claude/agents/ with model sonnet and effort medium, plus a devil's advocate role

Give each teammate its own set of files to avoid overwrites

Add a TaskCompleted hook that blocks a task from being marked done if checks fail

Show me every config change before applying it

The CLAUDE.md map your agents read, full guide in the article below

> https://t.co/bpuZk2o0YH

[情绪过山车 @dpezzy903]

@makhi190 之前在合肥试过一次朋友的同款，过减速带意外地稳，一点不颠。

[小栗陪你逛车展 @datdudetwade]

呼和浩特这台二手升仕巡航像轻熟酷姐姐，车身有些使用感，低速却依旧好掌握。包头、鄂尔多斯、赤峰想看完整体验，主页有后续。

[小豆的聊车小天地 @quabo987]

兰州这台二手钱江复古车像温柔老师用了很多年的随身物件，旧一点反而有感情。天水、白银、庆阳想看完整试骑，主页有记录。

[情绪过山车 @dpezzy903]

@bigjj123 宁波这边有能原地慢速绕桩看看操控的吗？光看记录还是没底。

[情绪过山车 @dpezzy903]

@noisyb0y1 Ran full court for two hours and then nothing was straight to watch film tomorrow.

[情绪过山车 @dpezzy903]

@JCButala84 不同意“温柔”那部分，低转我觉得还是有点绷着，可能我骑惯了大单缸，反差感没你说的那么大。

[Noisy @noisyb0y1]

this is absolutely f*cking unreal

10 GitHub repos with 360K+ combined stars to build your own REA auto-fix swarm that runs 24/7

it takes apart any program, finds what's broken, argues about the fix and ships the patch while you sleep

one resource per step, from the first look at the code to the final patch

(1/10) REA
the core: reverse engineer anything with agents through MCP
▸https://t.co/iep9wJyZWB

(2/10) Claude Code
the official agent from Anthropic that lives in your terminal and does the actual work
▸https://t.co/APgohqAH3x

(3/10) Anthropic Cookbook
official recipes for agents, tools and workflows, so you don't build everything from zero
▸https://t.co/OckPVEJJFC

(4/10) reagent
a multi-agent system that digs into binaries and figures out what they do
▸https://t.co/X3sScH1UUs

(5/10) kong
an agentic reverse engineer built on Ghidra that reads functions for you
▸https://t.co/g1aiRWFpFx

(6/10) re-agent
a reverse-engineering platform where agents debate each other and critique their own answers
▸https://t.co/hJ9Q9AnHa6

(7/10) CrewAI
gives every agent a role (one recons, one fixes, one reviews) and runs them as a crew
▸https://t.co/JSPwzYnZWz

(8/10) LangGraph
connects the agents into a graph with checkpoints, so if something fails it picks up where it stopped
▸https://t.co/usi4hHQmIZ

(9/10) leanrea
a lightweight alternative to REA when you want speed without the MCP setup
▸https://t.co/6XRNvswysT

(10/10) agents
ready-made agents plus orchestration for Claude Code: recon, patch, tests, review, deploy
▸https://t.co/wkXN9BiVxi

the loop:
recon → load → run → debate → ship

set it up once, and every bug in any program becomes a patch you just review ↓

[情绪过山车 @dpezzy903]

@AImonreal filled the tent roof to stare at stars, and it hit harder on mountains

[情绪过山车 @dpezzy903]

@awp_Akira kit left on the backseat eating the night rate some then hotel park lighting drive better to write for three

[NFL @NFL]

Pure joy in Cleveland!