X is so vibecoded that I got redirected to a profile called @Settings when I clicked on settings 🤣

I don't want to give names, but I am really getting tired of AI influencers who don't know shit and have no experience except on social media, yet speak garbage with so much authority just because they know how to farm for engagement

🚨 To use REA without losing your Claude or Codex account

Use GLM 5.3 or Kimi K3 through subscription or Openrouter, they're really good at reverse engineering

Do NOT use REA on your Claude or Codex account! You will get banned in a couple of hours

I felt like this needed to be said because this repo got really popular but Anthropic and OpenAI both forbid reverse engineering.

Hope this will prevent a wave of bans.

Opus 5.5 is great but can't use it's own sub-agents for security reviews

It often tells me that its security reviewer couldn't run because of the safeguards.

At this rate we will get the largest wave of insecure software in history

Gotta love how Anthropic gives us all this talk about "pacing the frontier", then starts running like there's no tomorrow

Who remembers Lovable?

In the old days they used to be a thing

It's funny how a month ago everyone was posting about Astra building some great visualizations/app and now all of this looks like the stone age already

GPT-6.1-Sol does very well when I make Opus 5.5 the boss

Essentially, I just plan with Opus and use GPT-6.1-Sol as an unlimited executor
Opus then reviews the work and gives feedback. Saves a lot of Claude usage while keeping work quality very high

Pro tip: use GPT6.1 sol with /fast. It is 2.5X faster and dirt-cheap

To the people hype posting "Opus 5.5 usage is unlimited"

Stop.

Anthropic is a company. As a company, their goal is to charge the customer as much as they are willing to pay and provide them a service that is as shitty as they can tolerate for a given price.
They take the temperature to determine this on socials among other sources.

If this hype gets large enough, they will actually decrease limits.

Why not do it if they can get away with it?

Grok bot squatting dot.com right after OpenAI's announcement of dot is the trolliest move 🤣 Who thought that up?

I see what Anthropic is trying to do here. It's not advertising of GLM 5.3 capabilities, it's more spreading fear about open models.
But it won't work, we need to thank ZAI and other model provides, because without them, without GLM 5.2, HuggingFace would have suffer severe damage from the OpenAI attack a few months ago.
We need more cyber capabilities in the Open to play both roles of red/blue teaming and increase security of our systems.

anthropic.com/research/glm-5…

Can't wait for GLM 5.4, they're gonna shit their pants
anthropic.com/research/glm-5…

I spent the last 3 days fixing/rewriting Astra slop with Opus. The difference is really that big.

In retrospect it really feels like Astra does not get what's important. Very capable autist

One agent file cuts Claude Code usage by 25% on a feature I built with 72 agents over 64 hours. Computed call by call on the real trace.

Unlike the main convo, sub-agents only keep their prompt cache for 5 minutes, even on a subscription. If one waits longer than that on a test t call pays for its whole context again. In that build it happened 200 times: 31% of the total cost.

The worst was one agent polling a slow job every 10 minutes. Each check cost ~20x a normal call (chart in the reply).

The fix is a sub-agent with a 1-hour cache. Save this as ~/.claude/agents/long-cache.md:

---
name: long-cache
description: General-purpose agent whose prompt cache lasts one hour instead of five minutes. Use it for agentic work that will sit idle more than five minutes between its own turns (waiting on long builds, test suites, visual checks, or its own subagents), especially once its context grows large, since every expired cache rewrites the whole context. For short or continuously active tasks use general-purpose: its cache writes cost less.
experimental:
cacheTtl: 1h
---
Work as a general-purpose agent on the task you're given.

Then ask Claude to use long-cache for anything that runs your slow commands.

Why not give every sub-agent a 1-hour cache? Its writes cost 60% more, and 53 of my 72 agents never waited 5 minutes. For them it's pure waste.

Numbers, from replaying all 11,267 API calls:
- average context at a rewrite: 563k tokens
- 1-hour cache on only the 18 agents that waited: -25.6%
- the agent in the image: 15 hours, 54 rewrites, 52 avoidable, half its cost

Details:
- the agent file needs Claude Code v2.1.248+, omptCacheTtl or CLAUDE_CODE_SUBAGENT_PROMPT_CACHE_TTL overrides it
- the 1-hour cache is ignored while a subscription is on usage credits
- on an API key, your main session is on 5 minutes too; set promptCacheTtl: "1h" for it
- replay method: API-price costs from every call's usage data; after a 5-60 min gap the prefix is read instead of rewritten, other writes cost 2x instead of 1.25x, gaps over 60 min still expire

Docs: code.claude.com/docs/en/prompt…
File: gist.github.com/shidenkai0/0e7…

Fable sits in a very weird place right now

less capable than Opus 5.5
5X more expensive
limited to 50% of your quota

My Codex weekly usage after the release of Opus 5.5

Guess that's why I see less people asking @thsottiaux for resets this week?

New models edit files through the shell instead of the write tool all the time.
It's completely unreadable in Pi, no diff

In my agent transcripts: 30% of Opus 5.5's edits in Claude Code, ~20+% for GPT-6, GLM 5.3, DeepSeek v4 flash and Kimi K3 in Pi.

I built an extension that it. It parses the command with tree-sitter, nothing is executed, ~0.2 ms per edit command/script.

Your thoughts @mitsuhiko?

Here's the before/after, link below
👇

Here's the repo, it's open source so feel free to try it out and mod it: github.com/shidenkai0/pi-…