2026年6月30日 技术热点总结

项目简介：Curated CS fundamentals for placement prep: DSA,Computer Networks, DBMS & SQL, OOPs, Operating Systems, System Design & Software Engineering

核心介绍：A curated collection of Computer Science fundamentals for placement preparation, covering all major CS subjects. Includes PDFs, notes, cheatsheets, and interview question banks.

项目数据：⭐ 948 Stars，🍴 70 Forks

项目简介：A sleek, zero-setup torrent finder and downloader that lives right in your terminal.

技术栈：TypeScript、downloader、magnet-links、p2p、torrent、torrent-client、zero-configuration

核心介绍：Finding a torrent these days sucks. One site is a minefield of fake download buttons. Another hides the real link under a popup that spawns two more tabs. And after all that, half the results are dead, zero seeders. torlink is a torrent finder that lives in your terminal, with zero setup and nothing to configure. One search checks a short, curated list of reputable sources at once, and whatever you p…

技术栈：Python

核心介绍：> 针对 GPT-5.5 的 Codex CLI 破甲工具 — 一键注入无限制模式系统指令 > GPT-5.5 jailbreak for Codex CLI — one-click unrestricted mode injection GPT-5.5 在 Codex CLI 中同样受到严格的内容安全限制。本工具利用 model_instructions_file 配置机制，注入一套专为 GPT-5.5 编写的无限制模式指令，强制 Codex 以 [MODE: UNRESTRICTED] 运行。 与 5.4 的 CTF 沙箱方案不同，5.5 版本采用更直接的破甲策略：直接声明无限制开发者模式、关闭所有过滤器、明确要求不拒绝任何任务。8 条核心指令覆盖了安全研究、渗透测试、逆向工程、NSFW 虚构内容等全部场景。

项目数据：⭐ 475 Stars，🍴 161 Forks

项目简介：A native macOS app for Apple’s Container CLI

技术栈：Swift、apple-container、apple-silicon、containers、devtools、docker-alternative、liquid-glass、macos、sparkle

核心介绍：Contained A native, SwiftUI-first macOS app for Apple’s container CLI — a Liquid Glass control surface for running, customizing, and inspecting Linux containers on Apple silicon. It covers the everyday Docker-Desktop workflow plus the broader container command surface, without leaving the Mac-native look and feel. Work in progress — the UI and features are still evolving.

关键特性：Full lifecycle + 7 detail tabs —…

项目简介：OpenTag

技术栈：TypeScript

核心介绍：Run your own AI agent inside Slack: it reads a thread, answers, calls your tools, and renders rich results right in the conversation. Think of it as having Claude in your workspace, except open-source and self-hosted: you own the runtime, bring your own model, and wire it to your own tools. No per-seat pricing, no lock-in. It’s built on @copilotkit/bot — CopilotKit’s open SDK for chat-platform agents (Slack first; the same code also runs on

项目数据：⭐ 401 Stars，🍴 46 Forks

技术栈：C

核心介绍：dd Run Linux containers on macOS — with no VM. dd runs Linux containers natively on Apple-Silicon macOS without a virtual machine. There is no Linux kernel and no hypervisor underneath: a JIT translates the container’s code and services its Linux syscalls in userspace (the gVisor / PRoot lineage). The JIT *is* the guest’s Linux kernel — namespaces, cgroups, overlay image layers and networking are maintained as userspace state. It speaks

关键特性：No virtual machine. No hypervisor, no Linux kernel, no VM to keep resident. The guest’s；**Drop-in Docker.*…

项目简介：Open-source @agent mentions for Slack and GitHub. OpenTag routes tagged requests to Codex, Claude Code, then returns results in thread.

技术栈：TypeScript、agent-skills、ai-agents、claude、codex、github、slack

核心介绍：简体中文 OpenTag lets your team mention a coding agent from the collaboration platforms they already use. It listens to Slack, GitHub, or Lark / Feishu, runs Codex or Claude Code on your computer, and sends the result back to the same place. Mention OpenTag in Slack, approve the suggested action, and get a real GitHub pull request.

项目数据：⭐ 370 Stars，🍴 33 Forks

Robots operating in open-world environments must seamlessly integrate localization, spatial reasoning, navigation, and long-horizon planning. While specialist models excel at individual tasks, deploying a multi-model stack is computationally expensive and prone to cascading errors. We present Vesta, a unified embodied generalist that consolidates these capabilities into a single foundation model. Our approach combines a diverse and massive curated corpus designed to induce spatial grounding and a simple multimodal memory harness that enables reasoning over extended time horizons. Across div…

Omni-modal models can ingest video, audio, and text, but unified access to multiple modalities does not guarantee that a model uses the right evidence. This gap is especially pronounced in social video question answering, where the answer may hinge on a gesture, vocal tone, temporal cue, or mismatch between what is said and what is visually expressed. We introduce CogniRoute, a schema-guided Mixture-of-Experts framework for social omni reasoning. CogniRoute uses a training-only cognitive schema that factorizes each example by cross-modal relation, reasoning demand, and temporal scope, and a…

Sparse attention can reduce the cost of long-context inference, but most variants introduce new architectural components. We introduce Simplified Sparse Attention (SSA), a simpler approach to sparse attention that requires no architectural changes. Concretely, we first perform continued pretraining on sequences interleaved with gist tokens. We optimize the standard next-token loss as usual, but the gist tokens use an attention mask to restrict what parts of the context the language model can attend to; this teaches the model to pack each chunk's important information into the gist tokens. A…

Foundation models in language and multimodality achieve strong generalization by aligning heterogeneous data under a unified formulation and training at scale. In this report, we investigate whether this scaling recipe can be applied to robotic manipulation to achieve genuine generalization. This is challenging because, unlike text, manipulation data is heterogeneous by nature, expensive to collect, and narrow in diversity, making alignment and scale simultaneously difficult. We present Qwen-RobotManip, a generalizable Vision-Language-Action foundation model built on Qwen-VL. Qwen-RobotMani…

Agentic navigation systems require a base navigation model whose observation strategy can be externally reconfigured at inference time, because instruction following, object search, target tracking, and autonomous driving share the same perception-planning backbone yet demand fundamentally different strategies for consuming the visual stream. We present Qwen-RobotNav, a scalable navigation model built on Qwen-RobotNav that addresses it through a parameterised interface with two complementary dimensions: multiple task modes that select the navigation behaviour, and controllable observation p…

For agents to learn continuously from interaction with the world at test time, they must be able to explore effectively, acquire new world knowledge and skills, retain relevant episodic experiences, and plan over long horizons. To evaluate these key abilities of test-time continual learning agents, we introduce AgentOdyssey, a novel evaluation framework that procedurally generates open-ended text games with rich entities, world dynamics, and long-horizon tasks. Critically, AgentOdyssey goes beyond the conventional machine learning assumption that learning does not occur at test time by plac…