Microagents
AI
self-improving agents
SHOW HN2023
How HN took it
4 positive, 10 neutral or asking, 0 mixed, 6 critical, out of 20 comments about the project. Thread score -0.09.
Idea: 3 for, 2 againstExecution: 1 for, 2 againstPrivacy: 0 for, 2 against
The thread, comment by comment
Each row is a top-level comment and its replies, in HN's order. Select a cell to open that comment on HN.
positive neutral or mixed critical about another commenter or something else the maker
What HN said
What people liked
Interesting analogy to Memento and prompt persistence
From 2 comments
The best analogy I can think of is that if you want your agent to accomplish something without persisting a long chat history, but instead use the agent to reorganize and change…
WiSaGaN on HNExcellent analogy
Hugsun on HN
Objections
Self-improving LLM loops are seen as unlikely to lead anywhere and may get stuck
From 2 comments
> What are you talking about? This doesn't describe the vast majority of human knowledge. I hope you don't mean to say that vast majority of knowledge is devoid of any consistent…
wg0 on HNThese and similar attempts are like those "perpetual motion machines" that captivated people's imagination for a long time. That the machine will generate energy and that would…
wg0 on HN
Safety concerns about self-replicating or runaway agent loops
From 3 comments
Add replication and you get virus like behavior. Not sure if this a good idea.
dukeofdoom on HNIf the author reads this: would highly recommend reading and thinking for a bit about safety :). Great project!…
bbor on HNIf you close that loop reliably enough it can literally go haywire and take over the planet. :P
3cats-in-a-coat on HN
Requests
Recommendations on books about agent-based systems
From 1 comment
Any recent book recommendations on developing agent based systems?
xBeats99 on HN
Questions
Can the agent eventually run the repo itself without the author?
From 1 comment
Is this anywhere close to create microagent that takes you (the author, "aymenfurter") out of the equation for this repo itself? Like to make yourself redundant and let the bot…
mattigames on HN
Does context size limit long histories?
From 1 comment
context size is still going to be an issue when history gets large, correct?
ilteris on HN
Alternatives named
aiconfig for prompt management
From 1 comment
Take a look at https://github.com/lastmile-ai/aiconfig for refactoring your prompt management - https://github.com/aymenfurter/microagents/tree/main/prompt_management
andrew_lastmile on HN
Magic Loops
From 1 comment
We do something similar with Magic Loops[0], but within the context of generating a single "loop" (automation). We've found that LLMs are pretty bad at prompting other LLMs,…
jumploops on HN