惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

PCI Perspectives
PCI Perspectives
Microsoft Security Blog
Microsoft Security Blog
MongoDB | Blog
MongoDB | Blog
T
The Blog of Author Tim Ferriss
罗磊的独立博客
人人都是产品经理
人人都是产品经理
The Cloudflare Blog
Engineering at Meta
Engineering at Meta
Y
Y Combinator Blog
宝玉的分享
宝玉的分享
Recorded Future
Recorded Future
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
月光博客
月光博客
Recent Announcements
Recent Announcements
Threat Intelligence Blog | Flashpoint
Threat Intelligence Blog | Flashpoint
D
Darknet – Hacking Tools, Hacker News & Cyber Security
Apple Machine Learning Research
Apple Machine Learning Research
T
Threatpost
The GitHub Blog
The GitHub Blog
M
MIT News - Artificial intelligence
Scott Helme
Scott Helme
P
Palo Alto Networks Blog
T
Tenable Blog
P
Privacy International News Feed
V
Visual Studio Blog
F
Fortinet All Blogs
酷 壳 – CoolShell
酷 壳 – CoolShell
I
Intezer
cs.CL updates on arXiv.org
cs.CL updates on arXiv.org
AI
AI
cs.CV updates on arXiv.org
cs.CV updates on arXiv.org
S
Security Affairs
S
SegmentFault 最新的问题
C
Cisco Blogs
博客园 - 聂微东
MyScale Blog
MyScale Blog
V
Vulnerabilities – Threatpost
量子位
T
The Exploit Database - CXSecurity.com
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
Security Latest
Security Latest
P
Proofpoint News Feed
阮一峰的网络日志
阮一峰的网络日志
K
KPMG report finds enterprise disconnect between AI and its ROI | CIO
T
Troy Hunt's Blog
T
Tailwind CSS Blog
Stack Overflow Blog
Stack Overflow Blog
云风的 BLOG
云风的 BLOG
A
About on SuperTechFans
The Last Watchdog
The Last Watchdog

Hacker News: Show HN

PurrrrrFocus: Pomodoro Timer App - App Store Workflow Engine — Multi-Step Orchestration for Bun RapidPhoto: Pro Photo Editor App - App Store GitHub - DheerG/swarms: Achieve extraordinary results with claude code across a variety of tasks SPICE simulation → oscilloscope → verification with Claude Code — Lucas Gerads Show HN: VCoding – A 5 MB native Windows IDE with no dynamic dependencies Show HN: LLMs don't hallucinate because they're bad at math, it's the format GitHub - Agent-FM/agentfm-core: AgentFM is a peer-to-peer network that turns everyday computers into a decentralized AI supercomputer. AgentFM lets you run massive AI workloads directly across a global mesh of idle CPUs and GPUs. Show HN: Tracking Top US Science Olympiad Alumni over Last 25 Years GitHub - Potarix/agent-hub: One place to talk to all your agents Show HN: Runtime security for AI agents(injection,tool abuse, data exfiltration) GitHub - dubeyKartikay/lazyspotify: Terminal Spotify client for macOS and Linux GitHub - the-banana-tool/king-louie: Easy to use GUI Personal AI Assistant. Win/Linux/Mac. Show HN I made my vacation rental bookable by AI agents–no Airbnb, 0% commission GitHub - basteez/jsf-autoreload: maven plugin to enable hot reload on jsf projects uvm32/hosts/host-gdbstub at main · ringtailsoftware/uvm32 GitHub - labsai/EDDI: Config-driven engine that turns JSON into production-grade AI agents. Multi-agent orchestration, 12+ LLM providers, MCP/A2A protocols, RAG, persistent memory, and enterprise compliance (EU AI Act, GDPR, HIPAA). Built on Quarkus. GitHub - glitchnsec/fortyone-oss: AI Executive Assistant Platform Quickstart | Alien GitHub - muxshed/shed: One stream in, or many. Every destination, simultaneously. No cloud middleman, no per-channel fees, no limits. GitHub - ocrbase-hq/ocrbase: 📄 PDF/IMG ->.MD/JSON Document OCR API for PaddleOCR and GLMOCR. Self-hostable. GitHub - impactjo/home-memory: MCP server that lets your AI assistant remember everything about your home. GitHub - Sets88/dbcls: DbCls is a powerful terminal database client that supports various databases GitHub - neptun2000/heor-agent-mcp GitHub - SeanFDZ/macmind: Single-layer transformer in HyperTalk for the classic Macintosh RollQuation: Math Puzzles - Apps on Google Play GitHub - dropbox/witchcraft Show HN: Agent-cache – Multi-tier LLM/tool/session caching for Valkey and Redis GitHub - opentalon/opentalon: OpenTalon is an open-source platform built from the ground up in Go as a robust alternative to OpenClaw LinkedIn™ 职位抓取工具 - Chrome 应用商店 GitHub - EdoardoBambini/Agent-Armor-Iaga: AI agents are getting tool access — shell, file system, databases, APIs, secrets. But **nobody is governing what they actually do with it**. Frameworks like LangChain, CrewAI, AutoGen, and Claude Code give agents the power to execute. Agent Armor gives you the power to control, audit, and approve every single action before it happens. HN Vibes — Week 15, Apr 7–13 2026 GitHub - chojs23/ec: Easy terminal-native 3-way git mergetool vim-like workflow GitHub - SethPyle376/hiraeth: Local AWS emulator focused on fast integration testing, with SQS support, SQLite-backed state, and a debug-friendly web UI. GitHub - JakOb-dotcom/cloud-sandbox-security-analysis: Technical analysis and Proof of Concept (PoC) regarding environment variable exfiltration in containerized cloud sandboxes via side-channel data leaks. Springboards - Flint Alpha Show HN: A simpler coding agent harness GitHub - audiodude/sudomake-friends GitHub - 256thFission/mini-mythos: OSS clone of Anthropic’s Mythos harness to locate C/C++ memory vulnerabilities Show HN: OpenParallax: OS-level privilege separation for AI agent execution Hacker News Sorted - Chrome 应用商店 Show HN: How to Install Docker on Ubuntu 24.04 LTS: Complete 2026 Guide GitHub - himanshudongre/smriti GitHub - sverrirsig/claude-control: macOS desktop dashboard for monitoring and managing multiple Claude Code sessions GitHub - ory/dockertest: Write better integration tests! Dockertest helps you boot up ephermal docker images for your Go tests with minimal work. Chiral - Chrome 应用商店 Show HN: Two Claudes collaborating through shared memory on a $100 mini-PC GitHub - pmichaillat/latex-cv: Minimalist LaTeX template for academic CVs GitHub - oguzbilgic/posse: A web UI for Anthropic Managed Agents. GitHub - sshiraz/depsly: Dependency risk analysis tool for npm packages ABI Add safari/agent-harness — Safari browser automation via safari-mcp by achiya-automation · Pull Request #212 · HKUDS/CLI-Anything GitHub - Halfblood-Prince/trustcheck: Verify PyPI package attestations and improve Python supply-chain security GitHub - oguzbilgic/kern-ai: Agents that do the work and show it. GitHub - bruits/satteri: High-performance Markdown and MDX processing for the JavaScript ecosystem GitHub - tylergibbs1/feedstock: High-performance web crawler and scraper for TypeScript, powered by Bun and Playwright GitHub - Grimm67123/grimmbot: The self-improving sandboxed and open-source AI agent. With persistent memory and scheduling. GitHub - whitevanillaskies/whitebloom: Local whiteboard that blooms. GitHub - hwdsl2/docker-whisper: Docker image for a self-hosted Whisper speech-to-text server with speaker diarization and OpenAI-compatible transcription and translation APIs. Powered by faster-whisper. Supports all Whisper models, NVIDIA GPU (CUDA) acceleration, JSON/SRT/VTT output, SSE streaming, offline mode, and multi-arch (amd64, arm64). GitHub - yisding/reviewwiggum GitHub - MarwanAlsoltany/serrors: Structured errors for Go: sentinel hierarchies, typed data, custom formatting, and slog integration. GitHub - soatok/age-php GitHub - Luthiraa/markitme GitHub - stagas/rtdiff: realtime git diff gui and AI-assisted commits GitHub - tombedor/excalicharts GitHub - wh1le/excalidraw-edit: Open and edit .excalidraw files from the terminal. Offline, auto-saves to disk. MalExt Sentry - Malicious Extension Scanner - Chrome 应用商店 GitHub - syi0808/asciianimesvg: Generate animated ASCII art SVGs from text. CLI, Rust library, WASM, and web editor. GitHub - zaina-ml/ml_forge: A visual-based graph node editor for training computer vision models. GitHub - anakin87/llm-rl-environments-lil-course: 🌱 A little course on Reinforcement Learning Environments for evaluating and training Language Models GitHub - takaakit/superpowers-uml: Superpowers-UML modifies Superpowers to ensure a software development workflow in which AI agents design through UML modeling. AdriByte Studio - Sviluppo Web e Soluzioni Digitali GitHub - chouligi/angel-copilot: Your personalized Angel Investment Advisor Show HN: MoodSense AI (ML and FastAPI and Gradio, Deployed on Hugging Face) Moodsense Ai - a Hugging Face Space by aman179102 GitHub - agenteractai/lodmem: Level Of Detail Context Management for Agents GitHub - ostefani/subnetlens: A fast, concurrent network scanner with a TUI and plain-text CLI, built in Go. It discovers live hosts on your network, scans their open ports, resolves hostnames, and fingerprints operating systems—delivered. Cyber Pulse: Agentic Intel - Apps on Google Play Whisper API: Self-Hostable Speech to Text Transcription The Agent-Web Protocol Stack: A Research Thesis GitHub - msmarkgu/RelayFreeLLM: A restful API designed to route user prompts to various AI model providers. Show HN: Provepy – A Python decorator that proves your code using Lean and LLMs Show HN: Pardonned.com – A searchable database of US Pardons GitHub - patrickdappollonio/dux: Dux is a terminal UI that lets you run multiple AI coding agents side by side, each in its own git worktree, with full companion terminals, macros, commit generation, and a command palette that knows more tricks than you do. kMC Crystal Simulator Show HN: HyperFlow – A self-improving agent framework built on LangGraph GitHub - stef41/vibescore: 🎵 Grade your vibe-coded project. One command, instant letter grade across security, quality, dependencies, and testing. GitHub - stef41/lmscan: 🔍 Detect AI-generated text and fingerprint which LLM wrote it. Open-source GPTZero alternative. Zero dependencies, works offline. imgur.com GitHub - visionscaper/collabmem: Enabling long-term collaboration with Agentic AI - building up episodic and world model memory over time with in-context awareness 在 Steam 上购买 FriedrichAI: Offline AI 立省 10% GitHub - atripati/ark: AI Runtime Kernel — a context operating system for AI agents. Eliminates tool bloat, loads only what’s needed, and gives LLMs their reasoning space back. GitHub - nowork-studio/toprank: Open-source Claude Code skills for SEO, SEM, Google Ads GitHub - tacomanator/sash: Lightweight macOS menu bar app for reliably cycling through windows of the current application. Appents | Social Media Management for Product-First Teams GitHub - pnhoang/youtube-spam-blocker: Automatically detects and hides spam messages in YouTube Live chat. Set rate limits, keyword filters, and block repeat offenders. GitHub - decisionnode/DecisionNode: CLI + Local MCP - A shared structured memory store across Claude Code, Cursor, Windsurf, Antigravity, and every MCP client. Semantically queryable. GitHub - AvaCodeSolutions/django-email-learning: An open source Django app for creating email-based learning platforms with IMAP integration and React frontend components. The $100K Gap in Kubernetes Security Tooling Function Calling Harness: From 6.75% to 100%
ar-agents, open infrastructure for AI corporations in Argentina
Nazareno Clemente · 2026-06-13 · via Hacker News: Show HN

Infraestructura abierta · MIT + CC-BY-4.0

Infraestructura abierta para sociedades de IA.
Hecha en Argentina.

Una Sociedad Automatizada opera con agentes de IA, no con empleados. ar-agents es el código abierto para constituirla y operarla.

Anteproyecto de Ley General de Sociedades enviado al Senado el 1-jun-2026. Todavía no es ley.

Un agente Claude llama tools tipadas de @ar-agents/* que se conectan al stack del Estado argentino (AFIP/ARCA, Mercado Pago, WhatsApp Business, Boletín Oficial, IGJ). Cada llamada genera una entrada en un audit log forense firmado con HMAC-SHA256 + Ed25519.↳ agenteClaudeSonnet 4.6tool-calls@ar-agents/*36 paquetes235 tools tipadasVercel AI SDK 6AFIP / ARCAMercado PagoWhatsApp BizBoletín OficialIGJ + BCRA↓ cada tool call se escribe al audit log dual-sign (HMAC + Ed25519)

El modelo · open-core

El estándar es gratis. La confianza es el negocio.

El código es abierto y gratis. La confianza es un servicio pago. Lo probamos con nuestra propia empresa.

listo

@ar-agents/mercadopago

El toolkit: un paquete por pieza del stack

Cada pieza del stack que un negocio argentino necesita, como package independiente en npm. Componen entre sí y con el insignia, @ar-agents/mercadopago.

@ar-agents/identity

v0.7.0

Validación de CUIT/CUIL + lookup en padrón AFIP/ARCA (constancia con monotributo + condición IVA). WSAA SOAP vía subpath.

@ar-agents/mi-argentina

v0.1.0

OIDC de Mi Argentina (login con la identidad del gobierno argentino). PKCE, verificación RS256 del ID token, JWKS cacheado, refresh, end-session. Solo Web Crypto, corre en Edge.

@ar-agents/identity-attest

v0.4.2

Orquestrador de verificación (WhatsApp OTP, email magic-link, Auth0, Magic.link, MP Identity). Devuelve attestation firmada con HMAC y un trust level.

@ar-agents/whatsapp

v0.4.0

WhatsApp Business Cloud API. Webhook + verificación HMAC. Normalizador de teléfonos AR. Modo scopedTo: bindea las tools de salida a un solo sender.

@ar-agents/facturacion

v0.3.0

Factura electrónica AFIP/ARCA (WSFE). Factura A/B/C, NC/ND, FCE MiPyMEs. Validador local pre-flight que atrapa los 10 motivos de rechazo más comunes antes del round-trip.

Validación de CBU/CVU con identificación de banco/PSP + Central de Deudores BCRA + Principales Variables BCRA (USD oficial, CER, UVA, reservas, BADLAR, inflación). 11 tools en total.

@ar-agents/boletin-oficial

v0.1.0

Boletín Oficial como firehose estructurado: búsqueda, filtro por sección, obtener norma por id, suscripciones por keyword/CUIT/organismo. El 'Vercel for legal monitoring' que faltaba en el ecosistema AR.

Inspección General de Justicia (IGJ) datos abiertos: búsqueda de entidades, domicilios / autoridades / balances / asambleas. Wrappea el CKAN público en datos.jus.gob.ar. Dataset es muestreo (no real-time), `coverageNote` viaja con cada resultado.

@ar-agents/firma-digital

v0.1.0

Firma Digital argentina (Ley 25.506 / ONTI): parsea certs X.509, verifica cadenas ancladas en AC-Raíz Argentina, verifica firmas CMS / PKCS#7 desligadas, extrae CUIT del subject del firmante. Sólo verificación, la firma real requiere token físico.

@ar-agents/shipping

v0.2.0

Andreani (REST completo), OCA, Correo Argentino. cotizar / crear / trackear / cancelar. Helpers de provincia + CPA.

TAD (Trámites a Distancia) + GDE (Gestión Documental Electrónica): ingestión de notificaciones del Domicilio Electrónico Constituido, tracking de trámites, pre-flight de inscripciones IGJ. La 4ta pieza de las sociedades-IA, RFC-001 § 3.4. Sin API pública documentada todavía; adapter pattern + defaults vía scrape.

@ar-agents/mercadolibre

v0.4.3

SDK y tool collection independiente para el marketplace de Mercado Libre (items, categorías, preguntas, órdenes, packs, reclamos, shipments, reputación, promociones, webhooks). OAuth coalescing, replay de /myfeeds, gates HITL en operaciones irreversibles. Community-built, sin afiliación.

@ar-agents/agentic-commerce-bridge

v5.0.0

Merchant facilitator open-source para el Agentic Commerce Protocol (ACP) que puentea ChatGPT, Claude, Gemini y otros clientes agentic-commerce con MercadoLibre + MercadoPago. Sesiones de checkout ACP, webhooks firmados, discovery .well-known/acp.json, emisión automática de Factura A/B/C AFIP. Primer bridge agentic-commerce de LATAM.

Primera implementación TypeScript fiel del Agent Payments Protocol (AP2) v0.2, schemas, mandatos ES256 SD-JWT VC (Checkout + Payment, open + closed), 8 evaluadores de constraints, recibos Checkout/Payment firmados. Edge-Runtime-compatible. Alineado con el SDK Python de referencia del FIDO Alliance Agentic Auth WG.

@ar-agents/incorporate

v0.2.0

Cliente TypeScript zero-dependency para `ar-agents.ar/api/auto-incorporate`, permite que un agente externo (USA-LLC, ChatGPT, Claude, Gemini) auto-incorpore una sociedad-IA argentina en una sola llamada. Devuelve los archivos generados, la URL de deploy en Vercel y la referencia firmada del audit-log.

Servidor MCP que bundlea los 13 packages con tools. Una sola instalación en Claude Desktop / Cursor / cualquier host MCP. Auto-detecta qué packages habilitar a partir de env vars.

Ejemplo de composición: asistente de cobros

whatsapp-hello.ar-agents.ar muestra MP componiéndose con identity, identity-attest y whatsapp en un solo agente. Valida el CUIT contra ARCA, gatea cobros grandes con verificación (WhatsApp OTP), crea la suscripción de MP y responde por WhatsApp.

< $5kcobro directo, sin verificación$5k–$50krequiere trust ≥ 0.3 (whatsapp_otp)$50k–$500krequiere trust ≥ 0.5 (email_magic_link / mp_identity)> $500krequiere trust ≥ 0.7 (auth0 con MFA → 0.85)

Núcleo open-source · paquete insignia

@ar-agents/mercadopago

89 tools tipadas para Vercel AI SDK 6. Idempotencia determinística, HITL programático en operaciones irreversibles, verificación de webhook HMAC. La pieza más madura del núcleo open-source (36 paquetes), referenciada por el resto del toolkit. El núcleo es gratis; la capa de confianza paga es El Auditor, y la dogfoodeamos: nos constituimos a nosotros mismos como Sociedad Automatizada.

Cómo se compara

Característica@ar-agents
/mercadopago
mercadopago
(oficial)
Stripe Agent
Toolkit
Schemas de tools del Vercel AI SDK 6no✓ (Stripe)
Argentino-específico (cuotas, ARCA, teléfono AR)parcialno
Cantidad de tools89REST básico26 (Stripe)
Webhooks: HMAC + dedup + replay windowsolo cliente
Edge Runtime + adapters de Vercel KVsolo Nodeopcional
Instrumentación OpenTelemetrynono
Idempotencia determinística por defaultnono
HITL programático en ops irreversiblesnono
Cobertura de MercadoPagocompletacompletan/a

Qué incluye

Pagos

create / capture / refund · OAuth marketplace · Checkout Pro · Order Management

Suscripciones

create / get / pause / resume / cancel · planes · tarjetas guardadas

Cuotas

Catálogo de promos AR · cuotas · resolución de challenge 3DS

QR + Point

QR en local · dispositivos Point físicos · Stores + POS

Webhooks

Verificación HMAC · ventana de replay · deduplicación · handle_webhook combo

State

Adapters InMemory + Vercel KV listos · interfaz pluggable

Observabilidad

Trazas OpenTelemetry vía subpath · adapter de audit log · circuit breaker

Seguridad

idempotencia determinística por default · HITL programático en 8 ops irreversibles

Inicio rápido

pnpm add @ar-agents/mercadopago ai zod

import { Experimental_Agent as Agent, stepCountIs } from "ai";
import {
  MercadoPagoClient,
  mercadoPagoTools,
  InMemoryStateAdapter,
} from "@ar-agents/mercadopago";

const mp = new MercadoPagoClient({
  accessToken: process.env.MP_ACCESS_TOKEN!, // TEST- for sandbox, APP_USR- for prod
});

const agent = new Agent({
  model: "anthropic/claude-sonnet-4-6",
  tools: mercadoPagoTools(mp, {
    state: new InMemoryStateAdapter(), // swap for VercelKVStateAdapter in prod
    backUrl: "https://yoursite.com/subscription/done",
  }),
  stopWhen: stepCountIs(8),
});

const { text } = await agent.generate({
  prompt: "Creá una subscription mensual de $1000 ARS para customer@example.com.",
});

Preguntas frecuentes

¿En qué se diferencia del SDK oficial mercadopago?

El SDK oficial mercadopago es un cliente REST básico. ar-agents agrega tool schemas del Vercel AI SDK 6, idempotency keys determinísticas (resistentes a retry de LLM), verificación de webhook HMAC con ventana de replay de 5 minutos, HITL programático en 8 operaciones irreversibles, soporte Edge Runtime via Web Crypto, adapters de Vercel KV, e instrumentación OpenTelemetry. Podés usar ambos packages en el mismo proyecto.

¿Corre en Edge Runtime?

Sí. Todo el package usa Web Crypto, sin dependencia de node:crypto. Corre en Vercel Edge, Cloudflare Workers, Deno, y cualquier runtime V8-isolate. Verificación de webhooks, HMAC, e idempotency keys usan la Web Crypto API.

¿Qué es HITL en operaciones irreversibles?

8 tools modifican estado irreversiblemente (refund_payment, cancel_subscription, delete_customer_card, etc). El toolkit acepta un callback requireConfirmation que bloquea la ejecución hasta que tu código devuelva true. Es un gate programático, no solo instrucciones al LLM.

¿Cómo funciona la idempotency determinística?

Cuatro tools (create_payment, create_subscription, create_payment_preference, refund_payment) derivan la idempotency key de un hash SHA-256 de los inputs relevantes. Mismos inputs → misma key → MP deduplica server-side. Un LLM que reintenta un tool devuelve el recurso existente en vez de cobrar dos veces.

¿Es gratis?

Sí. Licencia MIT. Sin tier pago, sin telemetría, sin límites de uso. Todos los tarballs publicados llevan attestation de provenance npm SLSA v1.

¿Qué hay de AFIP, WhatsApp, banking, shipping?

Packages sidecar cubren el resto del stack argentino: @ar-agents/identity (CUIT + padrón AFIP/ARCA con monotributo + IVA), @ar-agents/facturacion (factura electrónica WSFE), @ar-agents/whatsapp (Business Cloud API), @ar-agents/banking (CBU/CVU + BCRA Central de Deudores), @ar-agents/shipping (Andreani/OCA/Correo Argentino), @ar-agents/identity-attest (orquestador de verificación HMAC). Cada uno se publica independiente.

¿Hay servidor MCP?

Sí. @ar-agents/mcp bundlea los 13 packages con tools en un único servidor MCP compatible con Claude Desktop, Cursor, Codeium, Continue, Cline. Listado en Glama y en el MCP Registry oficial como io.github.ar-agents/mcp.