Muhammad Idrees @Idrees535

what is agent looping for the last two years we prompted agents one task at a time. that is starting to change instead of asking an agent to build the landing page and then driving every step yourself, you set up a loop that handles discovery, planning, the work, checking, and iterating until the goal is met looping is a setup you build. almost any agent harness can run it, it just depends on how you wire it up at its simplest, looping is one agent working on itself: > researches > drafts > checks the draft against a goal > fixes what is weak > runs that cycle again until the work clears the requirements you are not prompting each step anymore. the agent repeats the cycle for you the bigger version is a fleet looping. you give an orchestrator agent a goal, it breaks the goal into pieces, hands each piece to a specialist agent, and those specialists hand smaller jobs to their own subagents the whole tree keeps looping through discovery, planning, execution, and verification until the goal is met one agent looping is like a person redoing their own draft. a fleet looping is a whole team running a project end-to-end you create a goal, and the system runs the loop until it finishes within the reqs you set open and closed looping: OPEN LOOPING is exploratory. it still has conditions and a goal, but you give the agent or the fleet a wide space to move in. it can try different paths, discover things, build something you did not fully spec out this is the exciting end, it is what Peter and others are doing, and tbh it is where I want to spend more time the catch is cost, an open loop with real room to explore burns an insane amount of tokens. for the 90 percent of people without an unlimited budget it is not runnable yet, and pointed at projects with a loose standard it turns into a slop machine CLOSED LOOPING is bounded. a human designs the end-to-end path first: > clear goal > defined steps > an eval at each step > a point where it stops or hands back to you (and feeds back performance data) the agents still loop, but inside framework you built. it gets better every run because each pass feeds the next, and it runs on a normal budget because the path is tight. for most marketing work, closed is the one that pays off today. > the orchestrator owns the goal > the specialists own the steps > the subagents do the narrow work > an eval gate make sure its not slop

Peter Steinberger 🦞 @steipete

a month ago

Here’s your monthly reminder that you shouldn’t be prompting coding agents anymore. You should be designing loops that prompt your agents.

2K 1K 20K 8.5M 14K

Agents need better tools for reversing! I'm releasing declib (previously libbs), with a new CLI today that gives agents CLI access to 4 decompilers (IDA, Ghidra, Binja, angr), parity feature support to most MCP (12 features), and the ability to sync those changes across decs!

A harnessed LLM agent, clearly explained! Most people picture this as a model with tools bolted on. The real architecture inverts that relationship. The model itself is deliberately thin. Intelligence gets pushed outward, and the harness composes it at runtime. Three dimensions orbit the harness core: - 𝗠𝗲𝗺𝗼𝗿𝘆 holds the state a model shouldn't carry in weights or context. Working context, semantic knowledge, episodic experience, and personalized memory each have their own lifecycle. - 𝗦𝗸𝗶𝗹𝗹𝘀 hold procedural knowledge. This can cover operational procedures, decision heuristics, and normative constraints that specialize the general model per task. - 𝗣𝗿𝗼𝘁𝗼𝗰𝗼𝗹𝘀 hold the interaction contracts. Agent-to-user, agent-to-agent, and agent-to-tools are three distinct surfaces with their own failure modes. Between the core and these modules sit the mediators, like sandboxing, observability, compression, evaluation, approval loops, and sub-agent orchestration. They govern how the harness reaches out and how state flows back in. The useful question this framing unlocks is: for any new capability, where should it live? - Stable knowledge goes to memory - Learned playbooks go to skills - Communication contracts go to protocols - Loop governance goes to the mediators Harness design becomes a question of what to externalize, and how to mediate it. I'm building a minimal agent harness from scratch and will open-source it soon. In the meantime, my co-founder wrote an article about the anatomy of Agent Harness, covering the orchestration loop, tools, memory, context management, and everything else that transforms a stateless LLM into a capable agent. Read it below.

Akshay 🚀 @akshay_pachaar

3 months ago

x.com/i/article/2040…

82 564 3K 2.4M 8K

@androolloyd hl-node is talking

The rust is starting to give up its secrets. Interesting things happen when rust starts oxidizing. @34r7hm4n @mahal0z @Zardus @adamdoupe absolutely cooked. This thing is cursed in the best possible way. Everything is starting to click.

keyboard_arrow_left Previous keyboard_arrow_right Next

@androolloyd @mahal0z @34r7hm4n Cool, tell “the swarm” I said hi

@androolloyd @mahal0z @34r7hm4n Yeah this will be huge, already integrated it in my hl-node infra to track changes in each new release. HIP-4 is already on my radar

@mahal0z @34r7hm4n @HyperliquidX you guys need to buckle up now

@mahal0z @34r7hm4n @androolloyd this is awesome, trying it on hl-node and it is doing great, recovering symbols and structs which wasn’t possible previously with ghidra, you should give it a try

@34r7hm4n This looks amazing, just tried it bit and did much better than ghidra, cerbrus and bibary ninja. An mcp server of this would be great upgrade

@addyosmani will ai security be addressed at this harness layer or in model? form current trajectory it looks like frontiers are racing for performance to survive, while security is being ignored for performance and we already know that these next token predictors aren't safe anymore

x.com/i/article/2050…

THIS GUY JUST BUILT A DRONE THAT TRACKS TARGETS WITH A LASER USING CLAUDE > No robotics team > No engineering degree > No budget He used Claude Code to write the entire system in Python - a drone that locks onto a target, follows it with a laser and corrects its own aim automatically after every attempt. The AI gets more accurate the longer it runs because it learns from every single shot. He's still improving it and this is only the beginning. You have the same tool he used. The article below is about how to actually use it to its full potential.

Anatoli Kopadze @AnatoliKopadze

2 months ago

x.com/i/article/2051…

12 38 255 733K 679

@mphrediction I think the potential of fully on chain pms/outcome markets is huge as they will open new opportunities for risk hedgers. It will be the better direction leading to ARC framework's core philosophy medium.com/@idrees535/ada…

@androolloyd gist.github.com/idrees535/7bb1… Here is the gist, still not there, but hoping to run a devnet local validator with this. Should this work?

@androolloyd # Patch 11 change JNZ to unconditional JMP: 0245e65e (the JNZ LAB_0245e6a4) ---> JMP 0x0245e6a4 # Patch 12: 0245e974 (the JNZ LAB_0245ec92)--->JMP 0x0245ec92 0245e577 (the JNZ LAB_0245ec0b) --> JMP0x0245ec0b

@androolloyd doing binary patching to run a local validator node but it is trickier than I thought. Any ideas? # Patch 1 is_main_signer 04919de0 b8 01 00 00 00 MOV EAX, 0x1 04919de5 c3 RET

@androolloyd # patch 8 Stale State (Don't download always) 0245e68d → JMP 0x0245e6d5 # Patch 9 fix — change the JNC to unconditional JMP: 0245e6dd → JMP 0x0245e8c8 # Patch 10 force always take the success path: 04724ac2 (the JNZ LAB_04724d54)--->JMP 0x04724d54 Click OK

@androolloyd # Patch 4 computed hash ≠ quorum hash 1. Click on 0489c61a ---> NOP Same for 0489c61b, 0489c61c, 0489c61d, 0489c61e, 0489c61f

@androolloyd # Patch 3 Jailed Signers Patch FUN_04741050 to always return 1: 1. 04741050 (PUSH RBP)---> MOV EAX,1 3. 04741055 (PUSH R13 ---> RET → OK