Agents need control flow, not more prompts • Hacker News Board

Re: Agents need control flow, not more prompts

#141

Post by cogman10 » Thu, May 07, 2026, 9:43 PM UTC

DrewADesign wrote: post #109

827a wrote: post #100

1000% agree. I am increasingly hesitant to believe Anthropic's continual war drum of "build for the capabilities of future models, they'll get better". We've got a QA agent that needs to run through, say, 200 markdown files of requirements in a browser session. Its a cool system that has really helped improve our team's efficiency. For the longest time we tried everything to get a prompt like the following working: "…

I used to assume they pushed people into the prompt-only workflows because you’re paying them for the tokens, and not paying them for the scaffolding you built. However, I think that they’re really worried about is that a person needs to design and implement that stuff… It throws a wet blanket on their insistence that this will replace entire people in entire workflows or even projects, and I just don’t buy it. I do…

> However, I think that they’re really worried about is that a person needs to design and implement that stuff… It throws a wet blanket on their insistence that this will replace entire people in entire workflows or even projects, and I just don’t buy it.

I think you are on to something. But I also think this sort of system lends itself to not needing really good LLMs to do impressive things. I've noticed that the quality of a lot of these LLMs just gets worse the more datapoints they need to track. But, if you break it up into smaller and easier to consume chunks all the sudden you need a much less capable LLM to get results comparable or better than the SOTA.

Why pay extra money for Opus 4.7 when you could run Qwen 3.6 35b for free and get similar results?

Re: Agents need control flow, not more prompts

#142

Post by flowgrammer » Thu, May 07, 2026, 9:45 PM UTC

yogthos wrote: post #26

This was basically my realization as well. We are trying to get LLMs to write software the way humans do it, but they have a different set of strength and weaknesses. Structuring tooling around what LLMs actually do well seems like an obvious thing to do. I wrote about this in some detail here: https://yogthos.net/posts/2026-02-25-ai-at-scale.html

My experimentation with Verblets also concluded plain functions are the most logical harness for LLMs.

Re: Agents need control flow, not more prompts

#143

Post by moconnor » Thu, May 07, 2026, 9:46 PM UTC

“Flow” moves agents through a yaml flowchart of prompts and decisions. It’s working quite well for a couple of us in Tenstorrent, more to discover here though:

https://github.com/yieldthought/flow

Happily, 5.5 is good at writing and using it.

Re: Agents need control flow, not more prompts

#144

Post by reddit_clone » Thu, May 07, 2026, 9:47 PM UTC

sisve wrote: post #87

nixpulvis wrote: post #73

Earlier quoted context omitted.

My agents often write themselves scripts. Isn't that effectively what you're asking for? Prompting for scripts can also be a useful time and accuracy tactic when you know it'll be a good fit for it.

Yeah, the problem is that I do not think the agents is good at reusing scripts and stitching it together.At least for me it's recreating to much similar. I hope we will see platforms like windmill.dev find the optimal solution for this. I have not been able to test it enough. But have a platform that gives you some observability out of the box and protect secrets from llm is nice

I noticed that too. Unless you _ask_ for a script, they throw away the scripts they write.

They are particularly bad at complex multiline parsing. Writing all sorts of weird/crude python/awk scripts and getting confused in the process.

I wish they would use Perl6/Grammer or Haskell/Parsec or similar and write better parsing scripts.

Re: Agents need control flow, not more prompts

#145

Post by rdedev » Thu, May 07, 2026, 9:51 PM UTC

827a wrote: post #100

1000% agree. I am increasingly hesitant to believe Anthropic's continual war drum of "build for the capabilities of future models, they'll get better". We've got a QA agent that needs to run through, say, 200 markdown files of requirements in a browser session. Its a cool system that has really helped improve our team's efficiency. For the longest time we tried everything to get a prompt like the following working: "…

I had to create a hypothesis testing agent where it gets a query like "is manufacturing parameter x significantly different this month than last month" and have the agent follow a flowchart to run a statistical test and return the answer

At the time I had access to only 4o and there was no way to guarantee that the agent would follow the flowchart if I just mention it in its prompt. What I ended up wrapping the agent in a loop that kept feeding it the next step in the flowchart. In a way, a custom harness for the agent

Re: Agents need control flow, not more prompts

#146

Post by sharperguy » Thu, May 07, 2026, 9:52 PM UTC

827a wrote: post #100

1000% agree. I am increasingly hesitant to believe Anthropic's continual war drum of "build for the capabilities of future models, they'll get better". We've got a QA agent that needs to run through, say, 200 markdown files of requirements in a browser session. Its a cool system that has really helped improve our team's efficiency. For the longest time we tried everything to get a prompt like the following working: "…

So I wonder, if a more powerful agent harness could have the agent basically write and exectute its own deteministic code, which when executed, spawns sub agents for each of the subtasks?

So far we've seen agents spawn subagents directly, but that still means leaving the final flow control to the non-deterministic orchestrator model, and so your case is a perfect example of where it would probably fail.

Re: Agents need control flow, not more prompts

#147

Post by dnh44 » Thu, May 07, 2026, 9:54 PM UTC

swores wrote: post #119

mmis1000 wrote: post #112

Earlier quoted context omitted.

> This started breaking down after ~30 files. Codex's short context and todolist system combined somehow helps here though. Because of the frequent compact. The model was forced to recheck what todo list item has not done yet and what workflow skill it has to use. I used to left it for multi hour to do a big clean up and it finished without obvious issues.

Is Codex willing to do "multi hour" tasks when used with a ChatGPT Plus subscription, or does it need something more expensive like Pro?

I regularly get codex to do multi hour tasks with a single prompts I don't think thats a big deal anymore. But you don't want a single agent doing all the work. The root agent needs to delegate the work to sub agents. For example, a sub agent for context gathering, then one for planning, then one (or more) for implementation, then another for review. This way the root agent doesn't use up its context window and it just manages from a bird's eye view. I do have the $200 plan though.

Re: Agents need control flow, not more prompts

#148

Post by tonylucas » Thu, May 07, 2026, 9:55 PM UTC

sharperguy wrote: post #146

827a wrote: post #100

1000% agree. I am increasingly hesitant to believe Anthropic's continual war drum of "build for the capabilities of future models, they'll get better". We've got a QA agent that needs to run through, say, 200 markdown files of requirements in a browser session. Its a cool system that has really helped improve our team's efficiency. For the longest time we tried everything to get a prompt like the following working: "…

So I wonder, if a more powerful agent harness could have the agent basically write and exectute its own deteministic code, which when executed, spawns sub agents for each of the subtasks? So far we've seen agents spawn subagents directly, but that still means leaving the final flow control to the non-deterministic orchestrator model, and so your case is a perfect example of where it would probably fail.

I've been working on an integrated deterministic/agent integrated system for a few months now. It basically runs an AI step to build a plan, which biases towards deterministic steps as much as possible but escalates back to AI when it needs to (for AI only capabilities or deterministic failures) so effectively (when I perfect it, I'm about 90% there) it can bounce back and forward as needed with deterministic steps launching AI steps and AI steps launching deterministic steps as needed.

Probably not explaining it very well but I think it's pretty effective at reducing token usage.

Re: Agents need control flow, not more prompts

#149

Post by tonylucas » Thu, May 07, 2026, 9:57 PM UTC

sroussey wrote: post #102

827a wrote: post #100

1000% agree. I am increasingly hesitant to believe Anthropic's continual war drum of "build for the capabilities of future models, they'll get better". We've got a QA agent that needs to run through, say, 200 markdown files of requirements in a browser session. Its a cool system that has really helped improve our team's efficiency. For the longest time we tried everything to get a prompt like the following working: "…

I’m working on a hybrid system of old school task graph and ai agents and let them instantiate each other. I think others will do that eventually.

I'm working on something similar (won't link to it as don't want people to think I'm spamming) but if you want to compare notes happy to talk.

Re: Agents need control flow, not more prompts

#150

Post by dnh44 » Thu, May 07, 2026, 10:00 PM UTC

illwrks wrote: post #68

I’ve been building a small ‘agent’ using copilot at work, partly a learning exercise as well as testing it in a small use case. My personal opinion is that AI and agents are being misrepresented… The amount of setup, guidance and testing that’s required to create smarter version of a form is insane. At the moment my small test is: Compressed instructions (to fit within the 8k limit) 9 different types of policies to g…

To be honest Copilot really stinks and is really far from the sharp edge of what is possible these days.