Create

Sign in to ReadmeX

or

How we handle your data: Privacy Policy

ReadmeX

Your AI reading companion

What would you like to catch up on?

Pick a briefing, or ask your own question.

Your community and people briefings stay personal to you.

WorkBuddy

AI overview

Sign in and the AI will write an overview from our coverage.

Headlines · 189

  1. Looki builds a hardware-first path to personal AI agents

    Pinger reports that Looki is developing a chest-mounted, first-person device weighing about 31.8 grams to capture visual, audio and long-term experiences as context for personal agents. Its Loom temporal-perception model organizes continuous input into scenes, events and moments before supporting memory, insight and agent actions. Looki has opened Skills, plugins, APIs and MCP, and has connected Context with WorkBuddy, allowing other agents such as Claude and Codex to access relevant information. Founder Sun Yang says the next device will be smaller and use a lower-power platform, with audio kept online and visual input sampled at roughly one frame per second. In a 91-window sample from one user-day, Loom reduced label changes from 33 to 13; the report says this measures temporal stability rather than recognition accuracy.

    品玩 实时要闻 · 🔥 0
  2. How AI Agents Combine Models, Loops, Tools, and Context

    Juejin AI presents an AI agent as a model, an execution loop, tools, and context management working together, with the surrounding harness repeatedly invoking the model and returning tool results. It surveys planning patterns, low- and high-code implementation paths, multi-agent orchestration, persistence, and cross-machine collaboration, using WorkBuddy and Claude Code as examples. The article describes Claude Code as using ReAct-style execution with an optional Plan-and-Solve layer, parallel tool calls, and specialized subagents.

    掘金 · 人工智能 · 🔥 0
  3. Should Programmers Pivot to Agent Development or Full Stack?

    A Juejin AI analysis compares Agent development with full-stack engineering using job descriptions, salary data and practitioner views. It argues that full-stack capability provides the application and architecture foundation, while Agent development adds skills such as LLM integration, tool calling and agent frameworks; the piece advises programmers to validate demand through concrete projects and job applications rather than treating Agent work as an escape route.

    掘金 · 人工智能 · 🔥 0
  4. WorkBuddy and REA Workflow for Static Electron Analysis and Offline Replication

    In a first-person report by Xiaohu, WorkBuddy was connected to the open-source REA reverse-engineering tool to statically inspect two versions of a self-built Electron application without extracting its ASAR archives or running the app. The reported workflow included structural analysis, version comparison, replication of a local VoiceScribe tool using a Vosk offline speech model, revisions, and automated self-tests. An earlier run reported 21 passing end-to-end tests and five communication interfaces operating within 5.18 seconds; after revision, the workflow reported 27 chain self-tests. The capability and performance claims were not independently verified.

    掘金 · 人工智能 · 🔥 0
  5. Open-Source AI Workflow for Digital-Human Talking-Head Videos

    阿星 published an open-source workflow for producing digital-human talking-head videos with WorkBuddy and flashV4, covering topic selection, scripting, filming, effects packaging, cover creation and publishing. His reference setup combines interchangeable tools including LibTV MCP, Seedance 2.0, MiniMax, Wan 3.0 and Astra, with koubo-video-script and koubo-video-package skills handling scripting and packaging. The author says he tested the workflow himself but notes deformation or blur in the character, unstable cloned speech and occasional lip-sync misalignment.

    掘金 · 人工智能 · 🔥 0
  6. Doubao Work faces workflow and adoption hurdles in enterprise push

    A New Berry Daybreak investigation, authorized for republication by 36Kr AI and Cyzone Tech, examines ByteDance’s enterprise push for Doubao Work, including its Feishu integration, seat-plus-usage pricing, sales reorganization and announced opening to third-party models including Kimi 3.1 Pro, GLM5.3 and DeepSeek V4.1 Flash. User examples indicate that Doubao Work can save time on approvals, analysis and reporting, but may omit conditional steps or use outdated information. WorkBuddy, Qwen Office and Baidu Dazi are pursuing enterprise markets through different products and industries. ByteDance has not disclosed Doubao Work’s sales or contract figures.

    36氪 人工智能 · 🔥 0
  7. ApiGo 5.5 Adds Row- and Column-Level Data Permissions

    开源中国报道称,面向企业 AI 智能体的受控数据服务网关与 API、MCP 构建平台 ApiGo 已发布 5.5 版本,新增数据行级和列级权限。平台还宣称,可通过 WorkBuddy、千问办公等智能体以一句话创建并发布受控 API 与 MCP,并将 ERP、CRM、SRM 等系统数据封装为标准化 REST API 和 MCP 工具。

    开源中国 · 🔥 0
  8. WeCom AI Assistant Da Yuan Launches With Hunyuan Hy4

    WeCom's native AI assistant Da Yuan is reportedly fully available to enterprise users and integrated with Tencent Hunyuan Hy4. It can summarize chats, send messages, read and write documents and spreadsheets, compile project progress, run scheduled tasks, and coordinate with Agent bots such as WorkBuddy inside WeCom.

    36氪快讯 · 🔥 0
  9. Tencent Releases Hunyuan Hy4 With Multimodal and Computer-Use Capabilities

    Tencent released the Hunyuan Hy4 model on November 4 and said its weights would be open-sourced soon, while also introducing the lightweight Hy4 Flash. GeekPark reported that Hy4 ranked among the leading open models based on leaderboard results and demonstrated long-horizon agent workflows, visual understanding, visual coding, and computer operation in hands-on tests. Tencent’s claims about Hy4’s position among the strongest open models and its financial benchmark results remain vendor-supported claims. GeekPark also noted that a gap persists with leading closed models on difficult reasoning and engineering tasks.

    极客公园 · 🔥 0
  10. Tencent WorkBuddy brings AI agents into enterprise workflows

    A Tencent-provided article says its WorkBuddy office agent has handled more than 9 billion task requests, with token consumption rising over 50-fold after more than 100 version iterations. It cites August data from a third-party product ranking showing WorkBuddy first among Chinese desktop AI office agents by monthly active users, and describes deployments at Mengniu and Shanghai Jahwa. The article also outlines permission controls, sandboxing and execution tracing as governance measures for agents operating in enterprise workflows.

    量子位(原生 RSS) · 🔥 0
  11. One Author Tests WorkBuddy Browser Control in Five Workflows

    In a first-person hands-on report, author Xiaofan tested WorkBuddy’s browser-control feature across price comparison, job screening, event lookup, tutorial selection and form entry. The report says the feature supports Google Chrome, Microsoft Edge and QQ Browser, displays browser actions in real time and allows the user to stop them. Xiaofan found it useful for repetitive tasks but slower than manual clicking, and advised human review when submitting resumes or messages and when writing real data to production systems. These findings come from one author’s experience rather than a controlled evaluation.

    掘金 · 人工智能 · 🔥 0
  12. Open-Source AI Workflow Produces Two Whiteboard Animation Videos

    The Juejin AI article author says they used WorkBuddy to turn two narration scripts into an 83-second Long March route video and a 43-second Agent history video, both rendered at 1280×720 and 24fps. The seven-stage workflow synchronizes generated illustrations, handwriting, narration, subtitles, and camera movement, and is published as an MIT-licensed GitHub project containing 21 scripts, 28 tested parameters, examples, and an acceptance checklist. The author tested it on Windows 11 with Python 3.10 or later and FFmpeg, while noting that parameters may need adjustment on other systems.

    掘金 · 人工智能 · 🔥 0
  13. DeepSeek V4.1-Flash-1017 gets a hands-on agent test

    A Juejin AI article reviews DeepSeek's October 17 release of V4.1-Flash-1017, highlighting reported gains in Terminal-Bench, Automation-Bench and ProgramBench, plus vendor claims of lower task costs through token and agent-workflow optimization. The author reproduced DeepSeek's “wish-driven programming” racing-game example through DeepSeek Harness, finding that a one-line request produced a playable game with basic driving, tracks and timers, while a more detailed prompt improved its visuals and polish. The reported second run took 51 minutes and 14 seconds, consumed 60.1 million tokens alongside another smaller task and cost about 3 yuan.

    掘金 · 人工智能 · 🔥 0
  14. WorkBuddy Completes an 18-Step Storyboard-to-Video Test

    A hands-on report documents an 18-step WorkBuddy workflow that used three natural-language prompts to generate four 3D storyboard images and a 13.7 MB vertical MP4 with Chinese voiceover and subtitles. The roughly 14-minute process included installing edge-tts, inspecting frames, and repairing a subtitle-rendering defect. The author reported that the final MP4 did not play smoothly in the local player, leaving the complete audio, subtitle, and motion output unverified.

    掘金 · 人工智能 · 🔥 0
  15. agent-desktop 1.2.0 adds MCP support for background desktop control

    Vercel Labs released agent-desktop 1.2.0 with MCP support, according to 掘金 · 人工智能. The tool reads the operating system accessibility tree and assigns numbered UI controls so an AI can trigger them without screenshots or taking over the mouse and keyboard. In the author's WorkBuddy test, it made 55 error-free Calculator clicks and recorded a total of 1746 in Notepad, although actions were slow, consumed credits per step, and could not handle every interface in the background.

    掘金 · 人工智能 · 🔥 0
  16. HONOR details MagicOS November update with YOYO and cross-device improvements

    HONOR’s MagicOS team announced the November experience upgrade, as reported by IT之家. YOYO Suggestions and the Magic Capsule add shipping-code alerts, while YOYO Memory gains expiry reminders and smart form filling; HONOR Share adds WorkBuddy and Baidu Netdisk support, and Sogou Input Method receives a transparent mode. The rollout covers multiple Magic, WIN, folding, 600-series, GT Pro and MagicPad device lines.

    IT之家 AI · 🔥 0
  17. Deploying Qwen3.5 4B on a Legacy RX 580 Server

    A report from 掘金 · 人工智能 documents running Qwen3.5 4B with Ollama 0.40.14 on Windows 10 using an Intel Xeon E5-2670 v2, 32 GB of memory, and an 8 GB AMD RX 580. The author exposes the model to WorkBuddy and Dify through an OpenAI-compatible API, with the Xeon serving as fallback compute at a reported prefill speed of about 54–74 tokens per second. Because the Vulkan flash-attention path produced zero-filled output on longer prompts, the deployment disables flash attention and accepts slower generation for more predictable results.

    掘金 · 人工智能 · 🔥 0
  18. Tencent Cloud outlines enterprise AI MaaS stack

    At Tencent Cloud’s MaaS session, the company presented Hunyuan, TI-ONE, TokenHub and partner model practices as a stack spanning model development, deployment and business applications. It also described Hunyuan’s internal evaluation system, TI-ONE inference optimization on domestic accelerators and TokenHub’s launch of Xiaomi’s MiMo-V2.6 series. Tencent executives said model capabilities were improving, per-million-token call costs had fallen from yuan to jiao or fen, and agents were moving beyond chat interfaces toward planning and execution. The reported fivefold single-card throughput gain for Hy4 preview was measured in a specified domestic-accelerator test environment.

    InfoQ 中文 AI&大模型 · 🔥 0
  19. OpenViking launches connectors and knowledge compilation

    OpenViking, an agent context database from Volcengine, has launched connectors for Feishu documents, projects, and code repositories with scheduled synchronization. Its Compile command can turn imported project information into knowledge wikis, code graphs, and reporting materials; these capabilities come from the vendor and have not been independently tested.

    掘金 · 人工智能 · 🔥 0
  20. AI subscriptions hit Double 11 shelves as Token packs sell in livestreams

    TMTPost and 36Kr report that a Qwen Office personal basic monthly plan priced at 0.01 yuan sold nearly 100,000 copies in about 30 minutes during Li Jiaqi’s November 2 livestream, representing roughly 2 trillion tokens. Tmall and JD also promoted token packages and subscription plans during Double 11, with JD reporting that AI computers bundled with Token packages rose 100% in sales. The low-priced promotion generated little direct revenue, while retention and recurring paid conversion remain unproven.

    36氪 人工智能 · 🔥 0
  21. WorkBuddy Lets an Agent Modify a Kindle for Android and WeChat Reading

    In a firsthand account published by TMTPost, author 寻空的营销启示录 says WorkBuddy installed Android on a Kindle and then added WeChat Reading, handling the jailbreak, file copying, and commands while the author entered a URL. The article presents this as a consumer benefit of agentic AI: people can modify devices they already own without relying on paid specialists. TMTPost also cites broader examples involving Claude Code, CleanYourTV, and connected-home projects, though those examples were not independently verified in the supplied report.

    钛媒体 · 🔥 0
  22. WorkBuddy 5.8.4 unifies three connector paths around MCP

    A technical analysis of WorkBuddy 5.8.4 traces how its desktop capability bus, third-party connector packages, and user-written MCP servers are exposed to models through a unified connector interface. It says the paths retain separate credential storage and approval mechanisms, while describing controls such as loopback validation, route-specific tokens, context signatures, session allowlists, approval hooks, and configuration-hash-based consent. The author notes that the assessment is based on mechanism analysis rather than real OAuth testing, token theft, or local port scanning.

    掘金 · 人工智能 · 🔥 0
  23. WorkBuddy Update Loops and Edition Switching Reported

    A report describes two WorkBuddy updater problems: update prompts that reappear after installation or restart, and a domestic edition reportedly changing to the international edition after an update. The author’s explanation is that the updater compares version numbers and runs a silent installation. Proposed causes include a server-package version mismatch, files locked by running processes, a damaged cached package, and shared caches or installation directories between editions. The recommended recovery is to exit WorkBuddy and its updater, delete the updater cache, and reopen the app. The report says reinstallation may help when the installation is damaged or has international-edition remnants, but that cache deletion may not resolve a server-side version mismatch.

    掘金 · 人工智能 · 🔥 0
  24. Tencent Hunyuan Expands Hy4 Capacity for WorkBuddy

    Tencent Hunyuan and WorkBuddy said Hy4 reached many users' limits on its first day in WorkBuddy, prompting an urgent tenfold expansion of the Hy4 inference cluster and an increase in free usage allowances. IT之家 AI and 36氪快讯 reported the vendor announcement. Tencent said limits could still occur during peak periods, while Hy3's free use in WorkBuddy was extended to November 15, 2026, at 23:59.

    IT之家 AI · 🔥 0
  25. Tencent Expands WorkBuddy for Enterprise Agent Workflows

    Tencent announced new enterprise AI capabilities spanning WorkBuddy, office collaboration, knowledge and business-data access, security, runtime, storage, memory, and observability. The company also introduced the Buddy One integrated office suite and upgraded WorkBuddy Enterprise with system connectivity, governance, and cloud-hosted managed agents. GeekPark cited Suning’s use of the platform across 14 departments, while most broader capability and efficiency claims remain vendor-attributed.

    极客公园 · 🔥 0
  26. Tencent Hunyuan Hy4 benchmarks strongly as inference capacity expands

    Tencent Hunyuan Hy4 scored 35 on the Artificial Analysis Intelligence Index v5, ranking second among models from China-based labs, according to ArtificialAnlys. It posted 31% on FrontierCode 2.0 and 37% on Terminal-Bench 4.0, while its one-million-token context window is priced at $0.834 per million input tokens and $2.501 per million output tokens. Tencent also said it expanded Hy4's inference cluster tenfold and increased free user allowances following launch-day quota pressure, though the capacity figures were not independently verified.

    量子位(原生 RSS) · 🔥 0
  27. ApiGo v6.5.0 Released with JDK 25 Upgrade

    ApiGo v6.5.0 has been released with an upgrade to JDK 25, according to the report. The project description says the platform supports MCP connectivity to WorkBuddy, 千问办公, Trea Work, and 豆包. It also describes support for SQL and NoSQL data sources, including MySQL, Oracle, Dameng, TiDB, DolphinDB, Hive, and DuckDB, as well as online API publishing and management.

    开源中国 · 🔥 0
  28. OpenViking launches 3D STEP content-understanding model

    OpenViking launched a native 3D content-understanding model that imports and parses STEP files, generating orthogonal six-view and eight-axis side-view images, summaries, model details, and indexed PMI information for downstream agent reasoning. 掘金 · 人工智能 reports that OpenViking can support agent workflows including component retrieval, BOM preparation, and supply-change analysis through integrations such as Codex and Claude. OpenViking claims 100% information recovery, but no independent testing or reproduction was provided.

    掘金 · 人工智能 · 🔥 0
  29. WorkBuddy adds browser control with its own tab group and pointer

    In a firsthand Juejin article, author 苍何 reports that WorkBuddy recently gained a browser-control feature that can open pages, click and read content inside a browser where the user is already logged in. The work happens in a separate "WorkBuddy" tab group with its own pointer, so the user can keep using other windows, and tasks can also be dispatched from the mobile client to a chosen computer. The author shows it reading official docs, filling and confirming an enterprise OA expense form, reviewing a website as a novice user, and summarizing Bilibili comments, while noting that captchas, sliders and slow or changed pages may still require human takeover.

    掘金 · 人工智能 · 🔥 0
  30. China’s Internet Giants Reorganize Around AI Workbenches

    A NextFin News analysis republished by TMTPost reports that Tencent, Alibaba, and ByteDance are reorganizing around AI workbench agents and cross-application workflows. Tencent introduced Buddy One around the WorkBuddy agent desk, linked to Tencent Docs, meeting products, and knowledge tools, and shifted QClaw toward WorkBuddy while planning to shut QClaw in December. Alibaba merged the Wukong and MuleRun teams and publicly tested QoderWork under Tongyi Office. ByteDance grouped Feishu product teams with Doubao product teams. The report says internal enterprise systems would need to be adapted for agent use rather than establishing that the workbench agents already connect those systems.

    钛媒体 · 🔥 0

← PreviousNext →

Experience and discussion from the community

Share my WorkBuddy experienceAsk about WorkBuddy

Vibe-Research: An investment research workbench for China A-shares, US stocks, and HK stocks.

What problem it solves The project brings market monitoring, watchlists, portfolio records, research, news, backtesting, and report management into one workbench across China A-shares, US stocks, and HK stocks. It also keeps research records and imported materials on the user's computer while allowing research to be conducted with local AI subscriptions or compatible model APIs. Key features Unifi…

github.comhttps://github.com/simonlin1212/Vibe-Research

WorkBuddy launches the "Remote Control" feature

WorkBuddy launches the "Remote Control" feature Just open the app on your phone and connect to your computer to see a real-time view of the computer desktop and take control directly. Settings → Device Connection, enable Allow remote connections from devices using the same account. The computer must be turned on.

WorkBuddy