2026年11月5日 技术热点总结

项目简介：让你的命令行 Agent 在后台自己接活、自己跟进到底；平时不烦你，遇到拿不准的事才发卡片找你拍板。

技术栈：TypeScript、ai-agents、automation、claude-code、cli、codex、event-driven、mcp、nodejs

核心介绍：上手 · 场景模板 · 插件 · 内核 · 文档 平时我们用 Claude Code、Codex 这类命令行 Agent，大多是“人守着终端用”：你输一句话，盯着它跑，等它问你下一步；一旦关掉窗口或者人走开，事情就断了。 支持 macOS 和 Linux。运行前需要装好 Node 24、Git、curl，以及任意一个你已经登录好的命令行 Agent，不需要 GitHub 账号。也可以直接把安装说明丢给你的 Agent，让它替你装好。 > Windows 版：待定。 很多朋友在问 Windows 版，我都看到了。我现在人在外面出差，手边没有 Windows 电脑，没法真机开发和测试，所以现在还不能在 Windows 上用。这个周末（11 月 7、8 日）回家后，我会把它开发完，做完兼容性测试，再正式推出。进度会更新在 #17。 curl -fsSL https://raw.githubusercontent.com/KKKKhazix/agent-autoloo…

项目简介：Pika H3 Flex: video and audio generation inference

技术栈：Python

核心介绍：Pika H3 Flex Built for professional iteration. Six-step generation · Omni-reference · Native 480p / 768p · Up to 4K via Pika API Try it now · Model weights · Watch samples · Quickstart · Prompt guide · Pika API extensions Pika H3 Flex is a post-trained H3 model, designed to give professional creators the flexibility and speed to explore with precision. Optimized for greater control, it generates video with synchronized audio in six sampling steps and suppo

项目数据：⭐ 1,631 Stars，🍴 32 Forks

项目简介：Free Claude skill: viral animated story shorts (2D characters in a 3D world) with three.js + Remotion

技术栈：TypeScript

核心介绍：The free Claude skill from the video “How to Make Viral Animated Shorts Using Claude (Full Workflow)”. It makes 55-70 s vertical animated story shorts: 2D characters in a colourful 3D world, with ink outlines, a painted town, lens blur, gliding cameras, a march score and sound effects. Everything is rendered with three.js inside Remotion on your own PC, with no AI video generator. Proven on The Bridge That Fell for a Song, a 67 s short about th…

Reinforcement learning (RL) has shown promise for training multimodal agents to reason and use tools. However, it remains unclear when fine-grained credit assignment improves learning and when it fails to improve task performance. Video editing provides a useful testbed for this question, combining multimodal perception, long-horizon planning, and tool use with delayed outcome feedback. We construct four movie-trailer editing task types—Content Selection, Narrative Sequencing, Rough Cut, and Fine Cut—that reflect real-world editing workflows and span different levels of task complexity …

With model weights fixed, we measure how much of an entity-replacement response transfers between English factual questions, beyond whether answers change. We add the hidden-state change from two source questions to a target question about the same entity, comparing the predicted log-probability shift between two relations' published answers, given known answer prefixes, with the measured one. Hidden reuse injects the aligned change at 50% depth before the target's remaining layers; direct reuse adds source output-score changes. With other-relation sources, hidden reuse removes 45% and 49% …

Missing modality imputation aims to synthesize unobserved modalities from observed ones. Existing methods treat each modality as living on its own latent manifold, forcing the generator to learn a long, unstructured jump between them. We present SphereFlow, which takes a fundamentally different view: by projecting all modalities into a shared hyperspherical latent space via a frozen self-supervised encoder, cross-modal imputation reduces to short-range geometric transport between nearby points on the sphere. Realizing this idea with flow matching, however, exposes two geometric conflicts: t…

Why do language models fail on long inputs? The common answer is length. We show that, for retrieval, length is cheap and competition is not. On five open models (0.5B to 3B parameters, four families), neutral filler costs nothing: accuracy stays at 0.97 to 1.00, up to 32K tokens on Qwen2.5-1.5B. Statements in the format of the evidence, placed at the same positions, drop it to 0.16 to 0.33. A single statement that gives the queried key a rival value multiplies the loss by 9 to 110, and every additional rival adds to it. Inside the model, one quantity tracks the failure: the attention mass …

Heterogeneous embedded boards pair a Linux application processor with a real-time microcontroller (MCU), which suggests moving a task between the two domains when its timing degrades. We present QFabric, a prototype that treats cross-domain placement as guarded recovery from a violated empirical soft real-time contract: a bounded task is compiled for both domains, monitored in windows with hysteresis, switched only after semantic and capacity admission at a quiescent epoch boundary, and committed or rolled back after probation. On an Arduino UNO Q (Cortex-A53 Linux and a Cortex-M33 running …

Speculative decoding has been treated purely as a systems optimization that speeds up large language model decoding without changing the model behavior. Under this view, the growing effort to train better draft models has been focused solely on acceptance rate. We argue that this view does not hold true for long-horizon agent reasoning. Measuring frontier open-weight models on complex agentic tasks with different speculative decoding approaches, we found that numerical noise grows within verification windows, and the draft acceptance results of the thinking process can change the output dis…