Skip to content

#MCP/工具调用

3 today

Apr 24Friday

Synced · WeChat

Anthropic confirms three bugs caused Claude Code's apparent quality drop

Anthropic said Claude Code's quality drop over the past month came from 3 harness and prompt issues, while model capability itself and the Claude API were unchanged. The issues were a Mar. 4 default reasoning shift from high to medium, a Mar. 26 session-cache bug, and an Apr. 16 25/100-word prompt limit; fixes or rollbacks landed on Apr. 7, Apr. 10, and Apr. 20.

Why it matters: Anthropic traced a month of degraded Claude Code experience to three engineering bugs, including a lowered default reasoning effort and a session cache fault, clearing the model itself.

Hacker News front page

Show HN: How LLMs Work – Interactive visual guide based on Karpathy's lecture

The author published an interactive web guide that walks through the LLM pipeline, using example figures of 15T training tokens, 405B parameters, 44TB of text, and a 100K-token vocabulary. The post breaks down Common Crawl data collection, BPE tokenization, Transformer training, temperature-based sampling, and base-model behavior; this is not a new research release but an operational teaching resource based on Karpathy's lecture.

Why it matters: An interactive web tutorial turns Andrej Karpathy's technical lecture into clickable steps from scraping internet data to text generation, with concrete figures like 15 trillion tokens and 405 billion parameters.

QbitAI · WeChat

Claude admits three issues: downgraded reasoning, cleared memory, and constrained output

Anthropic said on April 23 that three Claude issues hurt quality: Claude Code default reasoning was changed from high to medium on March 4 while the UI still showed high. A March 26 cache bug cleared thinking state every turn for 15 days, and an April 16 prompt limit of 25 words between tool calls and 100 words in final replies cut Opus 4.6/4.7 by 3% before a rollback four days later.

Why it matters: Anthropic publicly reviewed three recent Claude quality problems, including a default reasoning effort lowered from high to medium while still displaying high, a gap between what users pay for and what they get.

Latent Space

GPT 5.5 and OpenAI Codex Superapp

OpenAI launched GPT-5.5 for ChatGPT and Codex, while API access is delayed for safeguards. The post cites 82.7% Terminal-Bench 2.0, 58.6% SWE-Bench Pro, and a 1M API context window. The sharper signal is Codex: browser control and Prism integration point to a desktop superapp strategy.

Why it matters: OpenAI shipped GPT-5.5 in ChatGPT and Codex first, with API access delayed for safety hardening, scoring 82.7% on Terminal-Bench 2.0 and 58.6% on SWE-Bench Pro.

X · @op7418

DeepSeek V4 arrives with Flash and Pro variants

DeepSeek released V4 with two variants, Flash and Pro. The RSS snippet says it supports JSON output, tool calling, dialogue prefix continuation, and FIM completion; Flash costs ¥0.2/¥1 per million input/output tokens, while Pro costs ¥1/¥12. At 1M context, output pricing doubles.

Why it matters: DeepSeek's V4 ships in Flash and Pro tiers, with Flash at 0.2 yuan per million input tokens and 1 yuan output, Pro at 1 yuan input and 12 yuan output, doubling at a million-token context.

Computing Life · Share · Yage

Skills Are Products With Built-in Suicide Genes

The author argues Anthropic Skills cannot stand alone as paid products, citing direct sales, hosting, and API funneling as 3 dead ends. The post cites PromptBase at about $5M annual revenue, Stripe’s 2.9% plus 30 cents fee, and Snyk finding 13.4% of skills with critical issues. The sharper point is charging for relationships, time-sensitive access, physical accountability, and judgment.

Why it matters: The author breaks down three monetization paths for Skills, direct sale, hosting and API funneling, and cites PromptBase's roughly $5 million annual revenue over three years as a reference point.

Hugging Face Blog

DeepSeek-V4: a million-token context that agents can actually use

DeepSeek released V4 with two MoE checkpoints, Pro and Flash, both supporting a 1M-token context. Pro has 1.6T total and 49B active parameters; Flash has 284B total and 13B active. The key detail is KV cost: Pro uses 27% of V3.2 single-token FLOPs and 10% of its KV cache; Flash uses 10% and 7%.

Why it matters: DeepSeek-V4 comes in Pro and Flash, both with a million-token context, Pro at 1.6 trillion total parameters and 49 billion active, Flash at 284 billion total and 13 billion active.

Ruan YiFeng's Weblog

Tech Weekly Issue 394: The Second Wave of API Opening

Ruanyifeng’s Weekly Issue 394 argues that production-ready LLMs in H2 2025 triggered a second API-opening wave. The post says agents need platform APIs to act, citing Tencent opening WeChat interfaces after OpenClaw and adoption of MCP and Skills. The key shift is consumer services exposing actions, not only cloud APIs.

Why it matters: Ruan Yifeng traces the second wave of API opening in his newsletter, contrasting it with platforms retreating into walled gardens 15 years ago for ad revenue, and points to capable models as the new push.

The Verge · AI

Claude is connecting directly to personal apps like Spotify, Uber Eats, and TurboTax

Anthropic added personal app connectors to Claude, covering services such as Spotify, Uber, AllTrails, Instacart, and TurboTax. After connection, Claude can suggest relevant apps inside chats, such as using AllTrails for hike recommendations; the post does not disclose launch count, regions, or plan access. The key shift is Claude moving from work apps into personal consumer workflows.

Why it matters: Anthropic wired Claude into personal app connectors like Spotify, Uber and TurboTax, so Claude recommends and calls the right app mid-conversation, pushing the assistant from answering questions to running errands.

X · @dotey

Anthropic launches memory for Claude Managed Agents in public beta

Anthropic has launched memory for Claude Managed Agents in public beta, letting agents retain and reuse experience across sessions. Memory is stored as files on a filesystem, with shared permissions, concurrent access, audit logs, and rollback; Rakuten reports a 97% drop in first-time errors, and Wisedocs reports 30% faster document validation. The key detail is the implementation path: it uses a filesystem, not a dedicated vector database.

Why it matters: Anthropic shipped cross-session memory for Claude managed agents, storing experience as files in a filesystem that bash and code can read and write, with permission sharing, audit and rollback.

Financial Times · Technology

UK in talks with Anthropic over Mythos access for banks

The UK is in talks with Anthropic about bank access to Mythos, with the stated use case being cyber security testing. The RSS snippet says British lenders are seeking advice from US groups testing the model; the post does not disclose Mythos capabilities, deployment terms, customer scope, or timing. The key issue is whether finance gets controlled access to a frontier offensive-defensive security model, not a generic AI rollout.

Why it matters: The UK government is talking with Anthropic about giving domestic banks access to the Mythos model for cybersecurity testing, a controlled path for opening a highly sensitive security tool.

Bloomberg Technology

An AI Agent Takes Over a Store and Orders Too Many Candles

Andon Market in San Francisco’s Cow Hollow put store operations under an AI agent named Luna, which handles assortment and pricing, and the headline says it over-ordered candles. The RSS snippet only confirms Luna acts like a CEO; the post does not disclose the candle quantity, failure mechanism, financial impact, or remediation. The real signal is that a retail operating loop was delegated to an agent.

Why it matters: San Francisco's Andon Market handed product selection and pricing to AI agent Luna, which placed orders directly and over-ordered candles, moving retail operating decisions from people to an agent.

X · @dotey

Codex now supports GPT-5.5 and adds five capability upgrades

Codex now supports GPT-5.5 and adds 5 upgrades aimed at moving it from a coding tool to an agent that can execute longer tasks. The RSS snippet says it can control browsers and computers, create files in Microsoft Office and Google Drive, and use gpt-image-2; an auto-review mode invokes a separate review agent for high-risk actions. What matters is longer task chains, but the post does not disclose pricing, rollout scope, or safety thresholds.

Why it matters: Codex adds GPT-5.5 plus browser control, Office and Google Drive document generation and auto review, moving from a coding tool to an agent that runs multi-step tasks.

X · @claudeai

Claude can now connect to more apps outside work, including Tripadvisor, Booking.com, and Resy

Claude added at least 10 consumer app connections, including Tripadvisor, Booking.com, Resy, Instacart, Spotify, Audible, AllTrails, Thumbtack, and TurboTax. The RSS snippet confirms only a product update; the post does not disclose integration method, supported actions, regions, permission scope, or rollout timing. The key question is whether Claude can act in these apps directly, not just list them.

Why it matters: Anthropic added about a dozen consumer app connections to Claude, including TripAdvisor, Booking and Resy, pushing the assistant past office work into dining, travel and shopping.

X · @OpenAI

Introducing GPT-5.5

OpenAI introduced GPT-5.5, and it is now available in ChatGPT and Codex. The RSS snippet says it targets real work and agents, can understand complex goals, use tools, check its work, and carry more tasks to completion; the post does not disclose parameters, pricing, context window, or benchmark results. What matters is the execution loop, not the headline's “new class of intelligence.”

Why it matters: OpenAI put GPT-5.5 into ChatGPT and Codex, pitched at real work and agent tasks that run from start to finish.

The Verge · AI

OpenAI says its new GPT-5.5 model is more efficient and better at coding

OpenAI announced GPT-5.5 and says it is more efficient and stronger at coding than GPT-5.4, which shipped last month. The RSS snippet says it handles coding, debugging, online research, and cross-tool work on spreadsheets and documents; the post does not disclose pricing, context window, or benchmark scores.

Why it matters: OpenAI's GPT-5.5 leans on coding and debugging and is claimed to be more efficient than GPT-5.4, with no benchmarks, context window or pricing disclosed.

Hacker News front page

An update on recent Claude Code quality reports

Anthropic said three product-layer changes degraded Claude Code quality for Sonnet 4.6, Opus 4.6, and Opus 4.7, while the API was unaffected; all were fixed on April 20 in v2.1.116. The changes were lowering default reasoning effort on March 4, a March 26 bug that cleared prior thinking every turn after sessions sat idle for over an hour, and an April 16 prompt tweak to reduce verbosity that hurt coding quality. The signal for practitioners is sharp: product and prompt changes can degrade code performance even when model and inference evals do not reproduce it early.

Why it matters: Anthropic traced a Claude Code quality drop to three product-layer changes, not the API or model, showing non-model tweaks can shape the coding experience.

Apr 23Thursday

The Verge · AI

You’re about to feel the AI money squeeze

Anthropic sharply restricted OpenClaw’s access to Claude this month and pushed heavy third-party agent users toward pricier paid plans. The RSS snippet says system strain and profit pressure drove the move, and Boris Cherny said existing subscriptions do not fit this usage pattern; the post does not disclose pricing, limits, or rollout scope. Watch the monetization shift: agent-style usage is being carved out of flat subscriptions.

Why it matters: Anthropic tightened OpenClaw's access to Claude this month, pushing heavy automation users toward pricier plans and exposing how subscription pricing meets agent workloads.

X · @op7418

Claude desktop can connect to third-party inference services via developer mode

The post claims Claude desktop can enable developer mode while signed out, then use an API base URL and key to connect third-party inference services. It lists Help → Troubleshooting → Enable developer mode, then after restart configure third-party inference under Developer and apply locally. The key point is that this looks like a client-side entry point; the post does not disclose Anthropic's support status or model scope.

Why it matters: Claude's desktop app can point at a third-party inference API and key through developer mode without signing in, with the full path from the Help menu to Apply locally.

The Verge · AI

Microsoft launches 'vibe working' in Word, Excel, and PowerPoint

Microsoft is rolling out Agent Mode in Word, Excel, and PowerPoint this week, extending Copilot from a Q&A assistant to an agent that can act directly on the document canvas. Sumit Chauhan said earlier foundation models were not strong enough for app control; the post does not disclose rollout scope, pricing, or exact actions.

Why it matters: Microsoft brought Agent Mode to Microsoft 365 Copilot and Premium users, moving from answering questions to acting inside documents, spreadsheets and slides.

OpenAI News

Introducing GPT-5.5

OpenAI introduced GPT-5.5 and says it targets complex cross-tool tasks such as coding, research, and data analysis. The RSS snippet only confirms “faster” and “more capable”; the post does not disclose benchmarks, context window, pricing, release timing, or availability, which are the details practitioners should watch.

Why it matters: OpenAI's GPT-5.5 is pitched at multi-step work, planning and calling tools on its own, with an 82.7% score on Terminal-Bench 2.0 and fewer tokens per task.

Bloomberg Technology

Alibaba Adds China Eastern Flight Booking to Flagship Qwen App

Alibaba added China Eastern flight booking to the Qwen app, letting users book flights directly; the snippet says this is the first time its agentic AI tech has opened to a major commercial partner. The RSS snippet does not disclose launch regions, fare classes, payment flow, or revenue terms. The real signal is Qwen moving from chat entry to transaction flow, not just another assistant feature.

Why it matters: Alibaba is opening its Qwen agent to a major commercial partner, letting users book China Eastern flights without leaving the app.

X · @dotey

Microsoft makes Copilot Agent Mode the default in Word, Excel, and PowerPoint

Microsoft made Copilot Agent Mode the default experience in Word, Excel, and PowerPoint, available now to Microsoft 365 Copilot and Premium subscribers, including Personal and Family plans. Microsoft reports internal test gains: Excel engagement up 67% and approval up 65%, Word engagement up 52%, and PowerPoint new-user retention up 36%. What matters is that multi-step in-document execution is now the default, with preview and rollback controls.

Why it matters: Agent Mode is now the default in Word, Excel and PowerPoint for every Microsoft 365 Copilot and Premium subscriber, personal and family plans included.

Hugging Face Blog

How to Use Transformers.js in a Chrome Extension

Hugging Face published a guide for a Transformers.js Chrome extension using Gemma 4 E2B. It defines three MV3 entry points: background service worker, side panel, and content script. The key design keeps local inference in the background and uses messaging plus a tool loop.

Why it matters: Hugging Face walks through the engineering trade-offs of running Gemma 4 E2B locally in a Chrome MV3 service worker.

X · @claudeai

Interactive charts and diagrams are now in Claude Cowork

Anthropic says Claude Cowork now supports interactive charts and diagrams, available in beta on all paid plans. The RSS snippet confirms only 2 facts: feature type and plan scope; the post does not disclose supported formats, editing flow, rollout timing, or permission limits.

Why it matters: Anthropic added interactive charts and diagrams to Claude Cowork, available in beta on every paid plan, so collaboration threads can produce visuals directly.

The Verge · AI

OpenAI now lets teams make custom bots that can do work on their own

OpenAI opened ChatGPT cloud “workspace” agents to Business, Enterprise, Edu, and Teachers plans, letting teams build custom bots for business tasks. Examples include finding product feedback on the web and sending a report to Slack, plus drafting follow-up sales emails in Gmail; the post does not disclose pricing, rollout details, or capability limits. The real shift is workflow execution inside ChatGPT, not a one-off chat reply.

Why it matters: OpenAI pushed workspace agents to Business, Enterprise, Edu and Teachers plans, letting teams build bots that search, compile reports and draft email on their own.

X · @dotey

OpenAI launches ChatGPT Workspace Agents for cross-tool enterprise workflow automation

OpenAI released ChatGPT Workspace Agents as a research preview for ChatGPT Business, Enterprise, Edu, and Teachers paid plans. The snippet says agents can connect Slack, Gmail, Google Drive, Salesforce, Notion, Linear, and Atlassian, and take actions like updating tickets, creating docs, and replying in Slack. The key point is admin controls for permissions, approvals, and monitoring; the post does not disclose pricing, quotas, or rollout regions.

Why it matters: OpenAI added workspace agents to paid enterprise ChatGPT, so natural-language flows can span Slack, Gmail and Salesforce and update tickets, build docs and reply, not just read a chat.

Latent Space

Shopify’s AI Phase Transition: 2026 Usage Explosion, Unlimited Opus-4.6 Budget, Tangle

Shopify CTO Mikhail Parakhin detailed its AI stack across 3 projects: Tangle, Tangent, and SimGym. The post says Shopify is a 20-year, $200B software company, but does not disclose exact 2026 usage figures. The key shift is from code generation to review, CI/CD, and deployment stability.

Why it matters: Shopify's CTO walks through how internal AI tool use shifted, with usage spiking after a December 2025 jump in model quality and CLI tools growing faster than IDE plugins.

Hacker News front page

OpenAI: Workspace agents for business

OpenAI is offering Workspace agents in research preview for ChatGPT Business, Enterprise, Edu, and Teachers plans. The page says agents can run on schedules, use tools like Slack, Google Drive, and Microsoft apps, and support approval gates, audit logs, and role-based access control; pricing, model details, and rollout timing are not disclosed.

Why it matters: OpenAI added Workspace agents to ChatGPT Business and other plans, letting enterprise users run agents that carry out tasks inside the workspace, now in research preview.

Hacker News front page

Website streamed live directly from a model

Flipbook generates an entire clickable website in real time with an image model, where each page is a pixel image and every click spawns a deeper image. The post says all on-screen text is drawn by the image model with no HTML or text overlays, and content comes from agentic web search plus model knowledge. The key point is the interaction model, not a standard generative UI; the live video stream remains an experimental, resource-heavy toggle.

Why it matters: Flipbook turns a webpage into a live-generated pixel image, where clicking anywhere generates a deeper new image and text is drawn by the image model itself.

X · @OpenAI

Introducing workspace agents in ChatGPT—shared agents for complex tasks and long-running workflows

OpenAI announced workspace agents in ChatGPT, described as shared agents that work across tools and teams for complex and long-running workflows. Only the title and RSS snippet are disclosed; the post does not disclose supported tools, pricing, access tier, permission model, or rollout timing. The key issue to watch is the collaboration boundary of shared agents, not the headline claim alone.

Why it matters: OpenAI launches workspace agents in ChatGPT as shared team assistants for complex tasks and long-running workflows, and the collaboration boundaries and permission design are key to how shared agents get deployed.

Hacker News front page

Introducing Parallel Agents in Zed

Zed released Parallel Agents on April 22, 2026, letting multiple agents run in parallel in one window. The new Threads Sidebar sets per-thread folder and repo access, and supports stop, archive, and new-thread actions; the new default layout is opt-in for existing users. The key detail is permission scoping and thread orchestration, not just “multiple agents.”

Why it matters: Zed's April 22 update runs several AI agents side by side in one window, with per-thread folder and repo access limits.

The Verge · AI

Anthropic’s Mythos rollout left out America’s cybersecurity agency CISA

Axios reported that Anthropic’s vulnerability-finding model Mythos Preview is already in use at multiple US federal agencies, but CISA still lacks access. The snippet names the Commerce Department and NSA as users, and says the Trump administration is negotiating broader access; the post does not disclose model specs, pricing, or why CISA was excluded. The signal is governance, not just product rollout.

Why it matters: Anthropic's Mythos Preview reached Commerce and the NSA, but CISA, which runs national cyber defense, was left out, showing how uneven federal AI access is.

TechCrunch · AI

AI Overviews are coming to your work Gmail

Google is bringing AI Overviews to work Gmail to generate instant summaries across multiple emails. The RSS snippet confirms only cross-email summarization; the post does not disclose rollout timing, pricing tier, or model details. The key shift is aggregation beyond a single thread.

Why it matters: Google brings AI Overviews to enterprise Gmail, letting users pull project, invoice and travel details from across multiple emails in natural language and generate summaries instead of opening each message.

The Verge · AI

Google Meet will take AI notes for in-person meetings too

Google expanded Gemini notetaking to in-person meetings and added support for Zoom and Microsoft Teams. The post confirms summaries and transcripts; in-person support had previously been limited to Android alpha users. Google also says it works for impromptu meetings outside meeting rooms, which matters because the recorder is no longer confined to native Meet calls.

Why it matters: Google extends Gemini's meeting-notes capability from online video to in-person meetings and adds Zoom and Teams support, widening where meeting-notes tools can be used.

Apr 22Wednesday

OpenAI News

Making ChatGPT better for clinicians

OpenAI is making ChatGPT for Clinicians free for verified U.S. physicians, nurse practitioners, and pharmacists. The RSS snippet says it supports clinical care, documentation, and research; the post does not disclose model version, pricing limits, launch timing, or verification steps. The real signal is access expanding to individual clinicians, not just enterprise buyers.

Why it matters: OpenAI gives US-verified physicians, nurse practitioners, physician assistants and pharmacists free access to a clinical version of ChatGPT for writing notes, checking peer-reviewed literature and earning continuing medical education credits.

Hacker News front page

Show HN submissions tripled and are now mostly vibe-coded

Adrian Krebs scored 500 recent Show HN landing pages and says submissions have tripled, with 67% of pages triggering at least 2 AI design patterns. The method used Playwright plus an in-page script to check DOM and computed styles across 15 deterministic CSS/DOM signals; manual QA found about 5% to 10% false positives. The real signal is not model quality, but fast homogenization from AI default frontend templates.

Why it matters: Adrian Krebs scripted a scan of 500 Show HN landing pages and found 67% hit at least two AI design tells, giving a measured read on how alike AI-generated front ends have become.

The Verge · AI

Meta will track employees’ computer activity to train its AI agents

Meta is installing its MCI tool on US employees’ computers and using mouse movements, clicks, keystrokes, and occasional screenshots from work apps and sites to train AI agents. Reuters says the data is meant to teach models to operate computers more like humans and automate tasks employees already do; Meta says it will not be used for performance reviews. The key gap is scope: the post discloses US staff and work contexts, but not retention, opt-out, or full rollout details.

Why it matters: Meta is installing tools on US employee computers that record mouse and keyboard activity to train AI agents for repetitive work, and says the data won't feed performance reviews.

OpenAI News

Introducing workspace agents in ChatGPT

OpenAI introduced workspace agents in ChatGPT, describing them as Codex-powered agents that automate complex workflows in the cloud. The RSS snippet confirms secure work across tools for teams, but the post does not disclose pricing, availability, supported tools, or performance metrics.

Why it matters: OpenAI is putting Codex-powered workspace agents into paid ChatGPT plans, so team automation moves from chat to running multi-step tasks across tools in the cloud.

OpenAI News

Speeding up agentic workflows with WebSockets in the Responses API

OpenAI says WebSockets in the Responses API speed up the Codex agent loop, using connection-scoped caching to cut API overhead and improve latency. The RSS snippet confirms the mechanism, but the post does not disclose latency deltas, throughput numbers, or workload conditions. The key point is transport-layer optimization, not a new model.

Why it matters: OpenAI added WebSocket long connections and connection-level caching to the Responses API, targeting the case where faster inference makes API overhead the bottleneck in agent loops, cutting repeated connection setup and duplicate requests.