1
Show HN: Rask – The .NET One Person Framework
专为独立开发者设计的.NET框架Rask，让你一个人搞定全栈开发。
Article URL: https://pal-tamas.github.io/rask/ Comments URL: https://news.ycombinator.com/item?id=49194562 Points: 1 # Comments: 0
2
Show HN: HN: Sandbox Your Agent with Runeward
给AI代理套上强制策略、审批和审计沙箱，开源治理新选择。
Article URL: https://github.com/Runewardd/runeward Comments URL: https://news.ycombinator.com/item?id=49180035 Points: 1 # Comments: 0
3
Show HN: I created a new Deep Learning Framework
纯Python+NumPy手搓深度学习框架picodl，pip一键安装源码全开放，想懂底层原理的开发者别错过。
I finally did it. I developed my own Deep Learning Framework from Scratch! I am calling it picodl. All written by hand, in pure Python, just using raw…
4
Show HN: Minimal LLM Post-Training Experiments on an 8GB GPU (SFT, DPO, GRPO)
只需8GB显存，就能跑通SFT、DPO、GRPO全流程，轻松理解DeepSeek-R1背后的推理训练奥秘。
Article URL: https://github.com/pochenai/nano-llm-posttraining Comments URL: https://news.ycombinator.com/item?id=49133851 Points: 6 # Comments: 0
5
AI Security Policy Should Assess Systems, Not Only Models
多智能体协同攻击框架揭示：评估AI安全不能只看模型，更要看系统。
arXiv:2605.09504v2 Announce Type: replace-cross Abstract: We present swarm-attack, an open-source adversarial testing framework in which multiple ligh…
6
Bound – A deterministic control harness for AI agents
开源的控制框架，让AI代理在迭代中知道何时停止，避免无限自嗨式的优化浪费资源。
Article URL: https://github.com/Danny-de-bree/bound Comments URL: https://news.ycombinator.com/item?id=48957765 Points: 2 # Comments: 1
7
Cybersecurity AI (CAI) Dataset
十四个月网络安全LLM轨迹数据集，直指专家操作经验才是性能提升关键。
Article URL: https://arxiv.org/abs/2605.28146 Comments URL: https://news.ycombinator.com/item?id=48841688 Points: 1 # Comments: 0
8
Show HN: TinyAgents – a Rust based recursive LLM harness
Rust编写的递归LLM框架，解决上下文衰减问题，让有效上下文突破原始限制。
Hey Guys, I would like to showcase Tiny Agents, which is an entirely Rust-based RLM. RLMs are a recent and fantastic new innovation on how LLMs can be…
9
Open handoff: Thought Tree, a markup/spec idea for modular LLM workflows
告别单条提示词的不可控，用结构化思维树让LLM工作流可执行、可检查，开源新框架值得一试
Article URL: https://github.com/RobertBateman/thoughttree-framework Comments URL: https://news.ycombinator.com/item?id=48701254 Points: 1 # Comments: …
10
Show HN: Core-1 – A distributed C++ AGI framework for trillion-parameter scale
分布式C++ AGI框架，专为万亿参数规模设计，开源可自部署，值得关注。
Article URL: https://github.com/Sarkar-AGI/Core-1 Comments URL: https://news.ycombinator.com/item?id=48705084 Points: 3 # Comments: 0
11
Show HN: LWDT – A Zephyr-inspired device-tree framework for ESP-IDF
把Zephyr的设备树玩法带进ESP-IDF，硬件描述更优雅，嵌入式开发值得一试。
Article URL: https://github.com/oldrev/lwdt Comments URL: https://news.ycombinator.com/item?id=48685336 Points: 3 # Comments: 0
12
graphlens: a polyglot code-analysis framework that turns your repo into a typed graph
将你的代码仓库转化为类型化图，GraphLens 让你的 AI 智能体比手动搜索快十倍。
graphlens: turn any repo into one typed graph — across Python, TypeScript, Go and Rust Every code-intelligence tool I've ever used falls into one of t…
13
Show HN: Pragmatiq – open-source framework for foundational models in banking
开源银行基础模型框架，助力金融AI落地，开发者可快速上手。
Article URL: https://github.com/dynamiq-ai/pragmatiq Comments URL: https://news.ycombinator.com/item?id=48601945 Points: 1 # Comments: 0
14
Show HN: Talos – Open-source WASM interpreter for Lean
开源框架Talos让WASM模块在Lean中实现形式化验证，为AI生成代码的可信执行提供数学证明基础。
At Cajal (YC W26) we’re excited to share Talos ( https://github.com/cajal-technologies/talos ), an open source framework for formal verification of We…
15
Show HN: Galdor – a Go LLM agent framework with built-in tracing and replay
Go语言LLM Agent框架，内置追踪与回放，开源且原生兼容OpenTelemetry，Agent开发利器。
Article URL: https://github.com/YasserCR/galdor Comments URL: https://news.ycombinator.com/item?id=48520360 Points: 3 # Comments: 0
16
Show HN: BeamWeaver – LangChain/DeepAgents-style agents and workflows for Elixir
为Elixir开发者带来LangChain/DeepAgents风格的智能代理与工作流框架，开源实现AI编排新可能。
Article URL: https://github.com/caudena/beam_weaver Comments URL: https://news.ycombinator.com/item?id=48503446 Points: 2 # Comments: 0
17
18
Sawtooth – An async, multi-tiered memory framework for LLM agents
专为 LLM 代理设计的异步分层内存框架，解决阻塞问题并优化记忆管理，开源可试！
Article URL: https://github.com/HtooTayZa/sawtooth-memory Comments URL: https://news.ycombinator.com/item?id=48422850 Points: 2 # Comments: 0
19
Anthropic's open-source framework for AI-powered vulnerability discovery
Anthropic开源AI漏洞发现框架，让自动化安全审计更强大。
Article URL: https://github.com/anthropics/defending-code-reference-harness Comments URL: https://news.ycombinator.com/item?id=48403980 Points: 16 # C…
20
微软发布 ASSERT 开源框架：让开发者更高效地评测 AI 智能体和应用行为
微软开源ASSERT框架，将自然语言规范转为可执行评估流程，助力AI智能体高效测试。
IT之家 6 月 3 日消息，微软今日宣布推出开源框架 ASSERT（Adaptive Spec-driven Scoring for Evaluation and Regression Testing，自适应规范驱动评分），旨在将自然语言写成的行为规范直接转换为可执行的评估流程。 据介绍，ASSE…