cabbit

24 | four years in AI research | now building with it daily @zscdao

[cabbit @cabbitar]

@zunosaid looks like the ultimate grokbot pack, thanks for sharing

[cabbit @cabbitar]

Opus + REA can rebuild any software feature without having a single line of original code

here is how to do it:

1. connect it to claude: run npx rea-agents setup

2. extract the rules: let rea inspect the feature and save a "spec" file

3. rebuild clean: start a fresh claude session with only that spec to rewrite it

4.test it: use /goal to test 3 intentionally broken versions. all 3 must fail to prove your new code is solid.

try running it, then read the full guide below:

[darkzodchi @zodchiii]

https://t.co/y7whCO2uJR

[cabbit @cabbitar]

How Anthropic engineers prompt Claude Code:

[cabbit @cabbitar]

How Anthropic engineers prompt Claude Code:

[cabbit @cabbitar]

https://t.co/9mobJryJ9p

[cabbit @cabbitar]

REA creator just released prompts that turn any AI agent into reverse-engineering lab

absurdly f*cking dangerous...

One paste it and your agents will:

• reverse engineer any paid app
• open Android, IPhone apps, Windows .exe and firmware
• take apart any app without its source code
• and that's just the start...

Send this to Claude Code, Codex or Cursor and thank me later.

Before you run it, read the article below:

[cabbit @cabbitar]

REA creator just released prompts that turn any AI agent into reverse-engineering lab

absurdly f*cking dangerous...

One paste it and your agents will:

• reverse engineer any paid app
• open Android, IPhone apps, Windows .exe and firmware
• take apart any app without its source code
• and that's just the start...

Send this to Claude Code, Codex or Cursor and thank me later.

Before you run it, read the article below:

[darkzodchi @zodchiii]

https://t.co/y7whCO2uJR

[cabbit @cabbitar]

fair point, but it's hard to fight slop when the algorithm rewards it 💀

[Nikita Bier @nikitabier]

Every social surface on the internet is in a state of constant entropy. Its default state is to decay and turn into slop. The only way a public space can endure is to be fighting tooth-and-nail every hour for its survival.

[cabbit @cabbitar]

REA + Opus 5.5 is really f*cking dangerous...

Send this single prompt to Claude and reverse-engineer any paid closed-source app

Then read the article below on how to become a REA engineer from scratch ↓

[codila @0xCodila]

REA is most dangerous thing on the internet right now...

It lets your AI agent reverse engineer any app, down to the binary

If you use it correctly, you can rebuild any feature you pay for, for FREE

In this article, I show you how: https://t.co/3P1FNXiXzl

[cabbit @cabbitar]

holy sh*t. I turned the Opus 5.5 cost mess into one prompt that finds where the money leaks

this will save you a lot of tokens…

drop in a run and it catches bloated context, late escalations and bad handoffs

checkpoint earlier → hand off less → lower $/pass

paste it into Claude Code and clean up the bill ↓

[cabbit @cabbitar]

https://t.co/9mobJryJ9p

[cabbit @cabbitar]

THIS CLAUDE.md FILE TURNS CLAUDE INTO A SENIOR ENGINEER

An Engineer from Claude Code shared the tricks and workflows that his team uses daily.

Most can be integrated directly into a CLAUDE.md for any project:
> Workflow Orchestration
> Sub-agent Strategy
> Self-Improvement Loop
> Verification Before Finalizing
> Automatic Error Correction
> Fundamental Principles

What's interesting is that it works as a cumulative system.

Every time you correct Claude, that correction can be recorded and turned into a rule for future tasks.

Over time, Claude learns from your feedback and can reduce the same errors again and again.

If you use AI for programming every day, having a well-structured CLAUDE.md can save you a ton of time.

Save this👇

[cabbit @cabbitar]

https://t.co/9mobJryJ9p

[cabbit @cabbitar]

ex. Anthropic engineer just shared the internal Opus 5.5 prompt for agent teams they used for Claude.

this is insane...

send it to your Opus 5.5 & Fable 5.5, and thank me later. Then read the full roadmap with examples below.

[cabbit @cabbitar]

https://t.co/9mobJryJ9p

[cabbit @cabbitar]

send this prompt to claude today , it will fix your entire life this weekend

[cabbit @cabbitar]

https://t.co/9mobJryJ9p

[cabbit @cabbitar]

@maestrooth so I just copy step 3 and send it to grok?

[cabbit @cabbitar]

Anthropic's Boris Cherny just shared a prompting system almost nobody uses

this is f*cking insane...

i applied it to my Opus 5.5 & Fable 5.5 prompt and it turned Claude into a 24/7 research machine.

send it to Claude, and thank me later. Then read the full Opus roadmap with examples below.

[cabbit @cabbitar]

https://t.co/9mobJryJ9p

[cabbit @cabbitar]

Official Anthropic tip for Claude Code: stop burning Opus 5.5 context on tasks Sonnet 5.5 can swarm, while Fable 5.1 sits idle

spin up an autonomous team with /teams claude --teammates ux,backend,adversary

paste the setup and this prompt into Claude Code below:

"Configure my Claude Code workspace for autonomous agent teams:
1. Audit ~/.claude/teams and .claude/teams for teammate profiles fitting ux, backend, and adversary > Generate definitions only for missing roles > Pin model: sonnet, effort: medium for workers, and model: fable for adversary > Assign each worker an isolated git worktree branch to eliminate file conflicts
2. Enable agent teams in ~/.claude/settings.json: > Set CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1 > Lock main session effortLevel to high
3. Initialize tasks.json in repo root with atomic file locking, allowing teammates to claim subtasks and resolve contracts directly over inter-agent messaging without coordinator intervention
4. Add a team rule to ~/.claude/CLAUDE.md: teammates must request an adversary security review before opening a pull request for the lead session Show every configuration diff first. Do not apply edits until confirmed"

Copy it and check out the full article below:

[cabbit @cabbitar]

https://t.co/9mobJryJ9p

[cabbit @cabbitar]

I'm sharing with you one of the best papers I've read on Harness Engineering.

It explains very well:

1) what a harness is
2) how current coding agents are built
3) what patterns repeat across Claude Code, Codex, Grok Bot, etc.
4) where the category is evolving toward
5) what they recommend if you want to build one

It all starts from a simple definition:
Agent = Model + Harness

Read the paper, then check out the article below:

[cabbit @cabbitar]

https://t.co/9mobJryJ9p

[cabbit @cabbitar]

Send this Jev prompt to your Opus 5.5

it installs Jev → audits your workflow → finds where you waste tokens, time and money

then moves repeatable decisions out of expensive LLM calls → upgrading your AI stack beyond 95% of people

Copy this now. Thank me later

[cabbit @cabbitar]

https://t.co/9mobJryJ9p

[cabbit @cabbitar]

@BreezeOg1 You have to at least give it a try fam!

[cabbit @cabbitar]

THIS CLAUDE.md FILE TURNS CLAUDE INTO A SENIOR ENGINEER

Boris Cherny, creator of Claude Code, shared the tricks and workflows that his team uses daily.

Most can be integrated directly into a CLAUDE.md for any project:
> Workflow Orchestration
> Sub-agent Strategy > Self-Improvement Loop
> Verification Before Finalizing
> Automatic Error Correction
> Fundamental Principles

What's interesting is that it works as a cumulative system.

Every time you correct Claude, that correction can be recorded and turned into a rule for future tasks.

Over time, Claude learns from your feedback and can reduce the same errors again and again.

If you use AI for programming every day, having a well-structured CLAUDE.md can save you a ton of time.

Save this, then check out the article below👇

[cabbit @cabbitar]

https://t.co/9mobJryJ9p

[cabbit @cabbitar]

Claude Code tip, and it's absolute free f*cking gold:

run Opus 5.5, Sonnet 5.5 and Fable 5.1 as one team and stop burning Opus tokens on routine work

the setup in one line: plan on high, delegate on medium, keep Fable on call

• who does what > Opus 5.5 on high - plans and ships the code > Sonnet 5.5 on medium - explorer reads code, worker edits and runs tests, researcher pulls docs > Fable 5.1 via /advisor fable - reads the whole session and speaks up only when it matters

• when Fable 5.1 steps in -> before a plan: is this the right approach? -> when an error repeats: am I digging in the wrong place? -> before "done": what did I miss?

Jev engineering takes it one layer lower: which file, which tool, retry or stop all go to Jev in under half a second, so the big models only see the real forks

paste this into Claude Code ↓

"Rebuild my Claude Code setup:

1. Find subagents in ~/.claude/agents and .claude/agents that fit explorer, worker and researcher. Draft only the missing ones. Set each to model: sonnet, effort: medium. List any that pin a different model and leave them

2. In ~/.claude/settings.json set effortLevel to high and advisorModel to fable.

3. Report anything that disables the advisor (CLAUDE_CODE_DISABLE_ADVISOR_TOOL, DISABLE_TELEMETRY, flag-fetching blockers) and CLAUDE_CODE_EFFORT_LEVEL.
Change nothing.

4. Add to ~/.claude/CLAUDE.md: consult the advisor before a large plan, when an error repeats, and before calling a long task done. Show every change as a diff. No edits until I say go."

↳ https://t.co/LPZ0BWfoiE
Give it to your Claude, then check out the full article below ↓ ↓

[cabbit @cabbitar]

https://t.co/9mobJryJ9p