Hacker News

Article URL: https://www.ibtimes.com/hackers-expose-discord-age-verification-system-issue-after-persona-frontend-code-left-wide-open-3797722

Comments URL: https://news.ycombinator.com/item?id=47107195

Points: 1

# Comments: 0

Built this because managing multiple AI providers was getting messy. ShuttleAI unifies Claude, GPT and other models behind one subscription API. Would love feedback from devs building agents or production AI workflows.

Comments URL: https://news.ycombinator.com/item?id=47107062

Points: 1

# Comments: 0

Article URL: https://opennhp.org/blog/the-internet-is-becoming-a-dark-forest.html

Comments URL: https://news.ycombinator.com/item?id=47107039

Points: 1

# Comments: 1

Article URL: https://github.com/llimllib/hatchat

Comments URL: https://news.ycombinator.com/item?id=47107026

Points: 1

# Comments: 0

Article URL: https://0ut0flin3.github.io/fr0g-protocol/frontend/

Comments URL: https://news.ycombinator.com/item?id=47107016

Points: 1

# Comments: 0

Article URL: https://en.wikipedia.org/wiki/The_Flying_Wallendas

Comments URL: https://news.ycombinator.com/item?id=47107005

Points: 1

# Comments: 0

My minimal LLM Agent (GPT4.1) just wrote it's own Agent Loop function.

Here is the backstory. I'm writing a tutorial: Zero to LLM Agent. Right now it is past 2 am here. I just wrote my 6th post. In the first five posts I wrote code that can talk to the OpenAI LLM. Then I gave it exactly one tool: a Python interactive environment (basically a Python REPL).

But, I used ipython and never actually implemented an agent loop. I defined an infer function infer(context, toolkit, evaluator) -> context, where context is just a simple list. The infer function then does the work. It takes the context, gives it to the LLM API, and performs the toolcall if needed.

Then I told the LLM Agent "The CWD is a Python project of an LLM Agent. Ignore the md file. Look only at the PY files in the CWD. Determine what is missing to make an LLM agent."

With a few more instructions by me, it implemented a perfectly reasonable agent loop function. The LLM Agent just improved it's own code.

Are we at a point where a bootstrap agent code can and will improve itself?

I would love to hear your thoughts on the matter.

Comments URL: https://news.ycombinator.com/item?id=47106996

Points: 1

# Comments: 0

Article URL: https://www.sambent.com/a-botnet-accidentally-destroyed-i2p-the-full-story/

Comments URL: https://news.ycombinator.com/item?id=47106985

Points: 1

# Comments: 0

Article URL: https://www.softwaremaxims.com/blog/hobbyist-gravity-well

Comments URL: https://news.ycombinator.com/item?id=47106976

Points: 1

# Comments: 0

Article URL: https://arxiv.org/abs/2602.17655

Comments URL: https://news.ycombinator.com/item?id=47106953

Points: 1

# Comments: 0

Article URL: https://www.os2museum.com/wp/the-history-of-a-security-hole/

Comments URL: https://news.ycombinator.com/item?id=47106937

Points: 1

# Comments: 1

Article URL: https://techxplore.com/news/2026-02-electrolyzer-plastic-syngas-ethylene-energy.html

Comments URL: https://news.ycombinator.com/item?id=47106918

Points: 1

# Comments: 0

Article URL: https://huggingface.co/blog/FINAL-Bench/metacognitive

Comments URL: https://news.ycombinator.com/item?id=47106917

Points: 1

# Comments: 0

Article URL: https://chromewebstore.google.com/detail/outliers/heogkfpbeagpoodininnfhdgjmpdalgj

Comments URL: https://news.ycombinator.com/item?id=47106908

Points: 1

# Comments: 1

Article URL: https://cloud.google.com/blog/topics/threat-intelligence/distillation-experimentation-integration-ai-adversarial-use

Comments URL: https://news.ycombinator.com/item?id=47106901

Points: 1

# Comments: 0

Over the last weeks I’ve been experimenting with a very narrow problem in the agent ecosystem:

We have marketplaces, orchestration frameworks, SDKs, and payment rails, but no minimal, framework-agnostic execution surface that lets heterogeneous agents actually coordinate work in a predictable way.

Most current setups tightly couple: - Execution logic - Economic logic - Framework internals

This makes interoperability brittle.

So I built a small experimental reference implementation that focuses on: - A middleware interception layer for tool-call validation - A minimal A2A JSON-RPC draft spec - An “economic hook” abstraction that can sit beneath orchestration or marketplace layers

Framework-agnostic design (CrewAI, AutoGen, custom stacks)

The goal is to solve coordination before economics.

This is still experimental, but I’d love feedback on: 1. Whether a standardized execution hook makes sense at the SDK boundary 2. If coordination should be separated strictly from routing/marketplace logic 3. What I might be missing architecturally

Curious to hear thoughts from others working on agent infra.

- Joaquin

Comments URL: https://news.ycombinator.com/item?id=47106895

Points: 1

# Comments: 0

Article URL: https://www.youtube.com/watch?v=oov8HDz3qVk

Comments URL: https://news.ycombinator.com/item?id=47106881

Points: 1

# Comments: 0

Article URL: https://nekoweb.org/explore

Comments URL: https://news.ycombinator.com/item?id=47106876

Points: 1

# Comments: 0

Article URL: https://reactdevelopment.substack.com/p/how-agentation-cut-claude-code-ui

Comments URL: https://news.ycombinator.com/item?id=47106870

Points: 1

# Comments: 0

Article URL: https://9to5mac.com/2026/02/20/apple-researchers-develop-on-device-ai-agent-that-interacts-with-apps-for-you/

Comments URL: https://news.ycombinator.com/item?id=47106866

Points: 2

# Comments: 0