Fast

The agent economy is in full swing 🔥⚡️ and @PiSquared has once again unveiled a fantastic new skill for agents, namely 𝐒𝐤𝐢𝐥𝐥 𝐎𝐩𝐭𝐢𝐦𝐢𝐳𝐞𝐫 – a tool that: → 𝑇𝑒𝑠𝑡𝑠 𝑎𝑔𝑒𝑛𝑡 𝑠𝑘𝑖𝑙𝑙𝑠 → 𝐶ℎ𝑒𝑐𝑘𝑠 ℎ𝑜𝑤 𝑑𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑡 𝐿𝐿𝑀𝑠 𝑎𝑐𝑡𝑢𝑎𝑙𝑙𝑦 𝑢𝑠𝑒 𝑡ℎ𝑒𝑚 → 𝐴𝑢𝑡𝑜𝑚𝑎𝑡𝑖𝑐𝑎𝑙𝑙𝑦 𝑖𝑚𝑝𝑟𝑜𝑣𝑒𝑠 𝑡ℎ𝑒 𝑖𝑛𝑠𝑡𝑟𝑢𝑐𝑡𝑖𝑜𝑛𝑠 𝑁𝑜 𝑔𝑢𝑒𝑠𝑠𝑤𝑜𝑟𝑘. 𝑁𝑜 𝑡𝑟𝑖𝑎𝑙 𝑎𝑛𝑑 𝑒𝑟𝑟𝑜𝑟. This is a tool for automatically evaluating and improving agent guidance, ensuring that tools are reliably used by different LLMs. 😎 Now it’s not just about creating an agent, but about teaching it to work correctly. Now agent skills aren’t just a case of “writing a prompt and hoping for the best”, but a measurable, optimisable system. 🧪 One skill → tested across multiple models 📊 Metrics → instead of subjective impressions 🔁 Auto-iteration → until it gets better This is the next step: Agents that don’t just exist, but improve. Welcome to optimised agent intelligence ⚡ #Fast #AI

The builders I respect in this community have all arrived at the same realization: Getting great AI output isn't prompting skill. It's staying organized aka "plan mode", working through a structure, not just dumping the problem into chat and hoping for the best. normalpowers packages that structure. Free, open source, works on Claude (Web, Desktop, Cowork, Code), OpenClaw, Hermes. Adapted from superpowers (the Claude Code plugin) for everyone, not just developers. Tried it on one of our GTM strategies last week. Came out with something actionable. If you've spent hours on an AI session and still got garbage, this is why. And it's a 10-minute fix.

@PiSquared Fast to the world

"You wasted all this time and tokens on nothing." every AI user has thought this. the problem isn't prompting, it's that AI answers before it thinks. you get a polished answer to the wrong question. every time. normalpowers fixes the front of the loop. free, open source. think superpowers, but for generalists.

@gesoosxbt yessir! at the top of the readme you have download buttons. moreover, if you give the GitHub link to your agent they should kinda know how to learn it if you ask it

@PiSquared everything i need to plug it here? x.com/PiSquared/stat…

@gesoosxbt absolutely. see which one you like best. or you can even talk to your agent to integrate normalpowers into your clarify workflow or viceversa

@PiSquared does it make sense to plug it in Hermes (that it has already the “Clarify” tool)?

@PiSquared 🦾🦾🦾

@gesoosxbt it's a plugin/skill that you can plug into your AI and it makes it ask more questions to understand your problem before rushing to solve it

@PiSquared what is this me no understand

@gesoosxbt what what

@PiSquared what

@HyderShav peak innovation is coming 👀 this is just a side project

@PiSquared Peak innovation by Fast ⏩

@Abyomiii what you're gonna build with it will be even more awesome!

@PiSquared This is awesome

@storka33 and normalpowers likes you, sir!

@PiSquared I like this

@1Idehen not even hard to ship this. we just needed it ourselves and wanted to share with the world

@PiSquared Devs keeps shipping

Observation after 1 week of running normalpowers on research and planning tasks: The core failure mode in AI-assisted knowledge work isn't model quality. LLMs are optimized to answer, not to structure inquiry first. Give a model an unstructured problem, get unstructured output. normalpowers adds the missing structure: 8 skills that make any LLM ask the right questions before generating output. Works on Claude (Web, Desktop, Cowork, Code), OpenClaw, Hermes. The right frame isn't "better prompting." It's "better epistemics for working with AI." Free. Open source. Worth 10 minutes of your time.

@sladalv use it and let us know how it helped!

@PiSquared Good help

@eunice1155 let us know what you build with it!

@PiSquared I need to test this. Keep building @PiSquared 🔥

@SirNicco your agent now asks questions then thinks FAST

@PiSquared Think FAST all the time.

@RosuGrigore if the AI is patient enough for @RosuGrigore , believe us when we say it will be patient enough for you

"Told you to first understand the problem well, and then answer." Each time I blamed the model ... "next version will be smarter". Truth is that good communication skills and asking the right questions are often more effective than intelligence alone. Both for humans and for AI.

@Ovcd27 massive kudos to Jesse, the main person behind superpowers! our devs were so stoked about superpowers that we decided that our non-devs need something similar as well!

@PiSquared if you are a developer, there's a chance you already know about superpowers. and if you don't, you're missing out! normalpowers is a fork of superpowers, switched up a bit so it's not focused only on software development github.com/obra/superpowe…

@aqeelerh feedback is welcome!

@PiSquared Lets put this to test