周报
2026-W36 周报精选：hermes-agent 仍在热榜；WAS Node Suite：ComfyUI 节点套件；ComfyTV：ComfyUI 电视项目
SwarmUI (formerly StableSwarmUI), A Modular Stable Diffusion Web-User-Interface, with an emphasis on making powertools easily accessible, high performance, and extensibility.
Multi-segment MiniMax H3 Director for official ComfyUI MiniMax-H3
WAS-NS Reborn; Tools for image processing, filters, masking, text, logic, numbers, latents, files, 3D scenes, and animation.
Clip chaining for MiniMax H3 in ComfyUI - motion and audio genuinely continue across joins
A ComfyUI custom node integration for local multi-engine multi-language Text-to-Speech and Voice Conversion. Supports: RVC, Echo-TTS, Qwen3-TTS, Cozy Voice 3, Step Audio EditX, IndexTTS-2, Chatterbox (classic and multili
ComfyTV — the canvas-based app that truly belongs to ComfyUI.
Custom ComfyUI nodes for film grain, color matching, and video enhancement.
🎨 Best DeepSeek Harness Design Plugin. The open-source Claude Design alternative. 🖥️ Local-first desktop app. 🖼️ Your coding agent becomes the design engine: prototypes, landing pages, dashboards, slides, images & video
See your agent think. Zero-config observability & governance for 26 AI agent runtimes: Claude Code, Cursor, OpenAI Codex, GitHub Copilot, Gemini CLI, Cline, OpenHands, Aider, Goose, OpenClaw, NVIDIA NemoClaw & more. Live
SmartGallery DAM is a local-first, browser-based Digital Asset Manager for ComfyUI. Generate without opening ComfyUI. Query your gallery with AI, in natural language. Advanced file manager, filter by prompt, model, LoRA,
A curated list of awesome resources for vibe coding
Gentle-AI configures the AI coding agents you already use: Claude Code, Cursor, OpenCode, Codex, Pi, and more. Choose persistent memory, Spec-Driven Development, curated skills, MCP servers, personas, and optional bounde
A ComfyUI custom node package for streamlined media loading and video pipeline assembly. Provides intuitive nodes that simplify media resource loading with user-friendly parameters, making it easier to build and configur
CRT-Nodes is a collection of custom nodes for ComfyUI.
The fastest local AI engine for Apple Silicon. 4.2x faster than Ollama, 0.08s cached TTFT, 100% tool calling. 17 tool parsers, prompt cache, reasoning separation, cloud routing. Drop-in OpenAI replacement. Works with Cla
自动生成电商带货短视频：上传商品图，AI 生成脚本、素材并合成视频，适配抖音、快手、小红书
Get web data for AI agents and LLMs - fast, efficient, and reliable with Rust
把 DevTools 接到 MCP。
Nous Research 的终端 Agent。今日 trending 仍在。
Comfy-Org 9 月 2 日打出 v0.34.3 正式版，节点工作流桌面/开源前端的补丁级发布。
开源 Agent 运行时/爪类编排（OpenClaw）。
docs(reference): 移除 voice-keychain 封面图，归档保持纯文本 Signed-off-by: Shinku
9 月 1–2 日还在合：自定义电池曲线、文档重排。ESP32-C3 可穿戴 Agent 徽章，站点 ai-passport.folotoy.cn。
This repository automatically updates a list of the top 100 repositories related to ComfyUI based on the number of stars on GitHub.
Omnigent is an open-source AI agent framework and meta-harness: orchestrate Claude Code, Codex, Cursor, Pi, and custom agents — swap harnesses without rewriting, enforce policies and sandboxing, and collaborate in real t
Documentation for ComfyUI
Governed execution layer for AI coding assistants: clarify intent, route capabilities, review evidence, verify results, and write back lessons across Claude Code, Codex, OpenClaw, and Cursor.
A prompt builder and media manager for MiniMax H3.
Local MCP server for ComfyUI — run your local ComfyUI from AI agents
为专业创作者而生的 ComfyUI 现代化管理平台
Video Delta Net hybrid attention for MiniMax-H3 as a native ComfyUI node.
Upscale long, high-resolution MiniMax H3 video on a VRAM-limited GPU in a single node.
Curated high-quality AI Agent Skills. Search, install, copy and share. Works with Claude Code, Cursor, OpenClaw, and other AI coding tools.
TypeScript multi-agent framework that runs in your own environment: consequential actions wait for approval and every run leaves a verifiable record. Describe the goal, not the graph. 13 built-in providers (Claude, OpenA
Seth's AI Tools: A Unity based front end that uses ComfyUI and LLMs to create stories, images, movies, quizzes and posters
Model-agnostic AI filmmaking workstation. 3D storyboard, ComfyUI local + Fal/Replicate cloud, multi-reference continuity, timeline export. Alternative to Storyboarder, LTX Studio.
给 Claude Code 的学术研究技能：检索、写作、审稿、修订一条龙。
开源编码 Agent CLI，可接云端和本地模型（OpenAI 兼容、Gemini、Ollama、Codex）。
Flovart 致力于补充画布自动化领域的缺失。它支持Claude Code、Codex等，让用户能够使用自己的第三方 API Key（如 seedance、banana2、gpt-image-2）来编排自定义视觉工作流，同时提供一种高级的、灵感来自lovart 的设计界面，方便用户日常使用。
OmO: Drop your tokens. Ultrawork. Done.
LoRA Pilot is an ultimate docker image for all Stable Diffusion LoRA trainers. Includes kohya_ss, diffusion pipes and TensorBoard for trainings and ComfyUI and InvokeAI for validation. Features shared models, modules, cu
Local-first AI token usage & cost tracker for 31 coding tools incl. Claude Code, Codex, Cursor, Gemini & DeepSeek Harness—with native apps. Never reads prompts.
A self-hosted blog for one person, with an MCP server so an AI agent can write and publish to it. One Bun process, two SQLite files, no cloud account in the path.
Comprehensive ComfyUI custom node suite featuring Smart Prompt with wildcards, a pipe ecosystem, universal type converters, image/video utilities, and workflow helpers. V3 API, Nodes 2.0 compatible.
Compare and generate AI videos across Sora, Veo, Kling, Seedance & more.
Personal fork: adds Wi-Fi OTA loader on top of FoloToy AI Passport. Original: github.com/FoloToy/ai-passport
Local-first AI Idea-to-video studio — FastAPI + SvelteKit + ComfyUI
Handle multiprompts and images within one run. Quick OutputLists from spreadsheet, JSON, multiline text, numberranges for sequential processing. Combinations of lists and prompts. Load any file with metadata and glob pat
ComfyUI for NVIDIA GPU in a Docker container with user configurable uid/gid
清华 MAIC 开源的多智能体课堂教学系统。
DIY 可穿戴。TRAE 是写玩法的 Agent，不当硬件。
六周内第三次 Flash：工作马 3.8 Flash 走编码/Agent，Cyber 走漏洞发现与补丁。入门价 $0.75/$3.75，年底翻倍。API、AI Studio、Gemini 应用可用。
You open the plugin catalog in Grok Bot for the first time. You search for X, find the plugin, and click it. A login screen opens in your local browser. You sign in, and you’re connected. You don’t need to get into the code of the system. You don’t need to install an MCP server J
This week, GitHub Copilot expands model choice and content protections, while VS Code adds new ways to manage agent sessions and get pull requests merge-ready. GitHub Copilot, general Claude Fable 5.1 is available to Copilot Pro+, Max, Business, and Enterprise users. Gemini 3.8 F
Claude code now appends every commit with this line, everyone is opted in automatically. This link is only accessible to the user so it doesn't make sense to be put in a commit. Claude-Session: https://cla
GPT-6 Astra from OpenAI is now available in GitHub Copilot. OpenAI’s latest general-purpose model, GPT-6 Astra, is designed for long-horizon, autonomous coding and agentic tasks. In our internal testing, GPT-6 Astra stood out for how it works, not just what it produces: it plans
TIL: Using Blender with coding agents on macOS I've been having fun with Blender in ChatGPT Codex on my Mac recently. Getting it to work with coding agents is really easy: install the full Mac application from blender.org and run a prompt like this: Use the already install /Appli
官方 prompting 指南，对着 5.1 改过。
Introducing Claude Fable 5.1 and Claude Mythos 5.1
Language-model judges now gate training data, score generations, and drive leaderboards. The judge is then a measurement instrument, resting on one rarely stated assumption: the same request, sent to the same model name, reads the same tomorrow. We audited that assumption in two
Language-model judges now gate training data, score generations, and drive leaderboards. The judge is then a measurement instrument, resting on one rarely stated assumption: the same request, sent to the same model name, reads the same tomorrow. We audited that assumption in two
The emergence of agentic AI and LLM-based multi-agent systems (MAS) presents unprecedented opportunities for automating complex tasks, while simultaneously raising critical concerns about the preservation of fundamental human-centered values, such as privacy, fairness, and safety
Modern AI agent harnesses expose lifecycle hooks that bind shell commands to runtime events such as session start, tool calls, and file edits. These commands run with host privileges yet ship as lifecycle-hook configuration and may fire at times the LLM never observes. We identif
Retail supply chain operations rely on coupled decision modules that must adapt as requirements evolve. LLMs offer a natural-language interface for this task, but existing methods primarily focus on individual optimization models. Extending them to heterogeneous decision pipeline
Repository-level software engineering benchmarks have significantly advanced the evaluation of coding agents, but existing benchmarks primarily measure whether generated patches pass functional tests and overlook review-derived acceptance constraints (review constraints) that oft
As large language models (LLMs) become increasingly capable, the long-term value of AI systems depends not only on solving individual requests, but also on transforming experience and accumulated knowledge into durable, reusable competence. We introduce SimSkill, a self-evolving
Computer-aided engineering (CAE) simulation is among the largest and most demanding areas of engineering, where setting up a solver such as OpenFOAM, FEniCS, or COMSOL takes real expertise. Large language model (LLM) agents promise to turn a natural-language request into a workin
Changelog 当天：Flash 进 Copilot。
It's not quite the "push button; get song" of Suno , but Roland's new Melody Flip tool marks the company's foray into generative AI music. Available as a plug-in for your digital audio workstation (DAW), Melody Flip offers around 250 "Palettes," which are essentially themed colle