Skip to content

MCP & tool use

How models connect to the outside world: the MCP ecosystem, function calling and tool integrations.

824 picksRelated topicsAgentsAI codingOpen source

Latest picks

521–540 of 824

Apr 27Monday

QbitAI · WeChat

Meshy tops 10M users and moves into 3D printing as ARR rises 14x

Meshy says it passed 10M registered users, reached $40M ARR, and grew 2025 revenue 14x year over year. Meshy Creative Lab supports keychain, magnet, and keycap design; physical ordering is not live yet. The key signal is print fit: 97% slice-pass rate in Bambu Studio across 75 tested models.

Why it matters: Meshy passed 10 million registered users and $40 million ARR, and its new Creative Lab generates models for small objects like keychains and fridge magnets, showing AI 3D content moving into physical consumer goods.

Hacker News front page

The Prompt API

Chrome’s docs describe the Prompt API for calling built-in AI inside the browser. The page links to session management and structured output docs; the captured body does not disclose model, context window, pricing, or rollout details.

Why it matters: Chrome opened its Prompt API to developers, letting web pages call the browser's built-in model to generate text without sending data to the cloud, with multi-turn sessions and structured output.

Synced · WeChat

From 99 Lines of Frozen Code to Meshy AI’s 3D Momentum in the West

Meshy AI released Meshy 6 and claims over 60% share in developed Western markets. The post says it has 10M+ users, $40M+ ARR, and 100M+ AI-generated 3D models in three years. The key signal is workflow fit: 37Games reports 30–40% less base sculpting work.

Why it matters: Meshy 6 ships with the company claiming over 60% of the US and European 3D AI market and a 30% to 40% cut in base sculpting work at 37 Interactive.

OpenAI News

An Open-Source Spec for Orchestration: Symphony

OpenAI released Symphony, an open-source spec for Codex orchestration. The RSS snippet says it turns issue trackers into always-on agent systems; the post does not disclose spec details, license, APIs, or benchmarks.

Why it matters: OpenAI open-sourced the Symphony orchestration spec, turning a Linear board into a dispatch point for coding agents where humans only review results; internal teams merged five times as many PRs in three weeks.

Apr 26Sunday

Hacker News front page

Agents Aren’t Coworkers, Embed Them in Your Software

Feldera co-founder Gerd Zellweger argues agents should be embedded in existing software, not treated as chatty coworkers. He lists 3 patterns: CLI, declarative specs, and Kubernetes-style reconciliation loops, then adds CDC streams for inserts, updates, and deletes. The key split: agents adapt logic, while the engine runs it continuously and emits precise changes.

Why it matters: Feldera co-founder Gerd Zellweger argues AI agents should be embedded in existing software to work quietly in the background, not treated as chatty coworkers.

Hacker News front page

Using Coding Assistance Tools to Revive Projects You Never Were Going to Finish

Matthew Brunelle used Claude Code with Opus 4.6 to rebuild a YouTube Music-to-OpenSubsonic connector, listing 6 setup steps. The stack used FastAPI, Pydantic, ytmusicapi, and yt-dlp, with Feishin logs used to fix .view suffix handling. The useful point: a clear spec plus human review beat one-shot generation.

Why it matters: The author used Claude Code to rebuild a shelved project into a YouTube Music to OpenSubsonic bridge, letting players like Feishin stream YouTube songs, with a spec-first, human-reviewed workflow worth copying.

Apr 25Saturday

Hacker News front page

Open-source memory layer Stash lets any AI agent do what Claude.ai and ChatGPT memory can do

Stash released an open-source persistent memory layer for AI agents, exposing 28 MCP tools and a 6-stage pipeline for long-term memory. The page says it uses PostgreSQL plus pgvector and hierarchical namespaces to separate user, project, and self memory. The real point is a portable memory layer, not the headline claim about matching ChatGPT or Claude.ai.

Why it matters: Stash ships an open-source memory layer on PostgreSQL and pgvector, with 28 MCP tools and a 6-step pipeline that adds fact extraction, a knowledge graph and contradiction checks beyond plain RAG.

Computing Life · Share · Yage

Anthropic lets Claude Cowork run rival models, a stranger move than it looks

Anthropic added an April 22–23 Claude Cowork switch for GPT-5.5, Gemini 3.1 Pro, DeepSeek V4, or local models. The post says third-party deployments have no Anthropic seat fee, and Bedrock, Vertex, and gateway prompts stay outside Anthropic. The key fight is runtime and control plane: AWS, Google, and Microsoft bet on Agent Registry, Apigee, and Entra Agent ID.

Why it matters: Anthropic lets Claude Cowork switch to GPT-5.5, Gemini 3.1 Pro or DeepSeek V4 for tasks, a month after cutting off third-party clients that piggybacked on subscriptions.

Computing Life · Share · Yage

TPU vs. CUDA: A Post-Cloud Next 2026 Assessment

Google announced TPU 8t/8i, TorchTPU, and an Anthropic deal at Cloud Next 2026; TPU 8i is slated for H2 2027 volume production. 8i has 288GB HBM, 8.6TB/s bandwidth, and 384MB SRAM; TorchTPU runs PyTorch on TPU, but the post says independent benchmarks are missing. The key crack is vLLM inference, while the author says TPU will not replace NVIDIA within 18-24 months.

Why it matters: At Cloud Next 2026 Google split its eighth-gen TPU into 8t for training and 8i for inference, and launched TorchTPU so PyTorch code runs directly without the old bridge layer.

Hacker News front page

Databases Were Not Designed for This

Arpit Bhayani argues agentic AI breaks four database assumptions: deterministic queries, human-reviewed writes, brief connections, and human-monitored failures. He proposes Postgres role timeouts of 5s and 10s, soft deletes, append-only logs, and idempotency keys. The key shift is treating agent_worker as an untrusted caller, not sizing pools like human-written apps.

Why it matters: Arpit Bhayani argues AI agents break four old database rules: fixed queries, human-reviewed writes, short-lived connections and someone watching for errors, and gives PostgreSQL timeout settings and soft-delete practices.

X · @dotey

Cursor 3 adds /multitask for parallel async sub-agents

Cursor 3 added /multitask and lets async sub-agents run in parallel. Queued tasks can also switch to parallel mode without waiting for the previous task to finish. The post does not disclose concurrency limits, resource usage, or failure rollback.

Why it matters: Cursor 3 adds a /multitask command that runs several async sub-agents at once, and queued tasks can be switched to parallel mode instead of waiting in line.

TechCrunch · AI

ComfyUI hits $500M valuation as creators seek more control over AI-generated media

ComfyUI raised $30 million at a $500 million valuation. The RSS snippet says its tools give creators more control over AI image, video, and audio generation; the post does not disclose investors, round stage, pricing, or release timing. The real signal is workflow control, not another model vendor.

Why it matters: ComfyUI raised $30 million at a $500 million valuation, betting that creators want to wire up node-based image and video workflows themselves for finer control over the output.

Hacker News front page

Could a Claude Code routine watch my finances?

Matt May used Claude Code routines with his Driggsby MCP server and Plaid to automate a daily finance email; he says the project took 2 months and about 75k lines of Rust. The post says the Gmail connector can only create drafts, so he added a restricted `email_me()` MCP tool that sends Markdown-only mail to a verified owner address. The practical angle is operability: routine behavior changes via prompt edits, and he already runs alerts on 7-day card anomalies and daily checking outflows over $500.

Why it matters: The author wired a self-written Driggsby MCP server to Plaid for bank and investment data, then used Claude Code routines to run a daily finance dashboard automatically.

TechCrunch · AI

Google to invest up to $40B in Anthropic in cash and compute

Google plans to invest up to $40B in Anthropic via cash and compute. The RSS snippet says it comes as AI rivals race for massive compute capacity and follows Anthropic’s limited release of the cybersecurity-focused Mythos model; the post does not disclose deal structure, timing, or compute allotment. Watch the compute tie-up, not just the headline dollar figure.

Why it matters: The deal's structure, $10B in cash up front and $30B tied to performance targets, partly paid in cloud compute, shows how tightly the two companies' compute is being bound together.

Apr 24Friday

Hacker News front page

Refuse to let your doctor record you

Emily M. Bender and Decca Muldowney give 9 reasons to refuse AI medical scribes. The tools record visits and draft chart notes, raising privacy, consent, automation-bias, and speech-recognition disparity risks. The key concern is clinics converting saved time into more visits.

Why it matters: Linguist Emily M. Bender and journalist Decca Muldowney give nine reasons patients should refuse AI scribes in the exam room, noting recordings and transcripts go to third-party firms.

Hacker News front page

Affirm Retooled Its Engineering Organization for Agentic Software Development in One Week

In February 2026, Affirm paused normal engineering work for one week and asked 800+ engineers to complete a full agentic workflow from ideation to submitted PR; it says over 60% of PRs are now agent-assisted. The post adds that 80%+ of engineers were weekly active users of AI dev tools by December 2025, and a nine-engineer group spent two weeks defining a default workflow around Claude Code, local-first development, and human checkpoints; the captured body does not fully disclose later implementation details or measured outcomes.

Why it matters: Affirm paused normal development for a week so 800-plus engineers could ship real requests to PRs with coding agents, and over 60% of the resulting PRs involved an agent.

Financial Times · Technology

Cohere and Aleph Alpha agree $20bn transatlantic AI tie-up

Cohere and Aleph Alpha agreed a $20bn transatlantic AI tie-up. The RSS snippet says they will focus on “sovereign” AI systems independent of the US and China. The post does not disclose the deal structure, funding split, product scope, or timeline.

Why it matters: Cohere and Aleph Alpha announced a $20 billion partnership to build sovereign AI systems independent of the US and China, putting the scale and stance of this transatlantic pairing on record.

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.