Buy A Modem

Show HN: Roe.md generate your own OpenClaw-like bot from a single Markdown file

Hi fellow HNers,I am guld and I created ROE.md (https://github.com/guld/ROE.md), a proof of concept single Markdown file that let's you raise your own custom personal OpenClaw-like AI assistant from scratch.With a very stripped-down feature-set of course. (For now. Still alpha)All you need is a little vibe code magic from your favorite AI coding tool. Select your favorite programming language (yes really) and hope that it has been grokked by the SOTA models. (I am lookin

Show HN: MicroGPT in 243 Lines – Demystifying the LLM Black Box

The release of microgpt by Andrej Karpathy is a foundational moment for AI transparency. In exactly 243 lines of pure, dependency-free Python, Karpathy has implemented the complete GPT algorithm from scratch. As a PhD scholar investigating AI and Blockchain, I see this as the ultimate tool for moving beyond the "black box" narrative of Large Language Models (LLMs).The Architecture of Simplicity Unlike modern frameworks that hide complexity behind optimized CUDA kernels, microgpt expose

Show HN: Mac apps are signed in. Why make an AI authenticate too?

Hi HN, I built an open-source (MIT) AI agent for macOS that talks directly to Mail, Calendar, Reminders, Notes, and Safari through AppleScript — no OAuth flows, no API gateways, no passwords to store.The premise is simple: your Mac apps are already signed in. Why should an AI assistant make you re-authenticate everything through a separate integration layer? Why do I need to create an Azure app to get an OAuth token for my Office 365 emails? Why do I need a Google PubSub setup?I needed something

Show HN: MarginDash – See which AI customers are profitable

I built MarginDash because I couldn't answer a basic question: which of my customers actually make me money after AI costs? My Stripe dashboard showed revenue going up. My OpenAI bills showed costs going up faster. I had no way to connect the two at the customer level.MarginDash is a few line SDK integration (TypeScript, Python, or REST) that tracks model usage per customer and connects it to revenue — either through Stripe sync or by passing revenue directly in the API call. You get a per-

Show HN: ToMusic – Turn text prompts into original songs with AI (iOS)

Hey HN,I built ToMusic, an iOS app that turns text descriptions into full songs (lyrics, vocals, and instrumentals) in seconds.How it works: You describe what kind of song you want (genre, mood, tempo, vocal style) or paste your own lyrics, and the app generates a complete, original song. It also works for pure instrumentals like background music, beats, and soundtracks.It supports 150+ genres (Pop, Jazz, EDM, Classical, Hip Hop, Metal...), 30+ moods, and 5 different AI models each with differen

Show HN: SatGate – Budget enforcement proxy for MCP tool calls (L402/macaroons)

I built an open-source MCP proxy that enforces per-tool budgets on AI agent tool calls.The problem: MCP gives agents tool access with no economic controls. An agent stuck in a retry loop against a $0.10/call API burns real money until someone kills it. Auth says "who" and "what" — nothing says "how much."How it works: SatGate MCP Proxy sits between your agent and upstream MCP servers. Transparent relay — agents don't know it's there until they hit the

Show HN: CoChat MCP – Let your team review what your coding agent is building

I built an MCP server that connects coding agents (Claude Code, Cursor, OpenCode, Codex) to a collaborative workspace where your team and other AI models can review what the agent is planning.The problem: When Claude Code creates an implementation plan, it lives in your terminal session. Nobody else sees it until it becomes a PR. If you want GPT to check the architecture or a teammate to flag issues, you're copy-pasting between windows.This MCP server fixes that. When your agent creates a p

Show HN: PolyMCP – Orchestrate AI agents across Python tools and MCP servers

Hi everyone,I am Vincenzo and i’m working on PolyMCP, an open-source framework that not only exposes Python functions as AI-callable MCP tools but also lets you orchestrate agents across multiple MCP servers.The idea: instead of rewriting code or wrapping every function with a special SDK, you can: 1. Publish your existing Python functions as MCP tools automatically 2. Spin up a UnifiedPolyAgent that coordinates multiple MCP servers 3. Ask your agent to perform complex workflows spanning diff

Show HN: Flemma – a Neovim plugin where the .chat buffer is the conversation

Hey HN, I posted Flemma back in October 2025 with no context. Since then I've shipped >100 commits and used it daily as my primary AI workspace so I figured a proper update was due.The core idea: a .chat file IS the conversation. No SQLite, no JSON logs, no shadow state. What you see in the buffer is exactly what the model receives. Edit an assistant reply to fix a hallucination, delete a tangent, fork by duplicating the file - it all works because there's nothing to fall out of syn

Show HN: The biggest achievement of my life so far

Hello everyone,I have always loved coding and in the couple I was thinking of making an open source project and it turned out to be awesome I hope you guys like it.I present Explore Singapore which I created as an open-source intelligence engine to execute retrieval-augmented generation (RAG) on Singapore's public policy documents and legal statutes and historical archives.The objective required building a domain-specific search engine which enables LLM systems to decrease errors by using g

Show HN: Gigacode – Use OpenCode's UI with Claude Code/Codex/Amp

Gigacode is an experimental, just-for-fun project that makes OpenCode's TUI + web + SDK work with Claude Code, Codex, and Amp.It's not a fork of OpenCode. Instead, it implements the OpenCode protocol and just runs `opencode attach` to the server that converts API calls to the underlying agents.We build this to scratch our itch of being able to rapidly switch between coding agents based on the task at hand. For example, we find that:- Claude Code is the best executor & fast iterator

Show HN: Sediment – Local semantic memory for AI agents (Rust, single binary)

I've been increasingly relying on AI coding assistants. I recently had my first child, and my coding hours look different now. I prompt between feedings, sketch out ideas while he naps, and pick up where I left off later. AI lets me stay productive in fragmented time. But every session starts from zero.Claude doesn't remember the product roadmap we outlined last week. It doesn't know the design decisions we already made. It forgot the feature spec we iterated on across three sessi

Seedance 2.0 preview: The best video model of 2026, outperforming Sora 2

ByteDance quietly shipped Seedance 2.0. The interesting part isn't the usual text-to-video upgrade — it's the reference/conditioning system.What's different from the typical T2V model:Accepts 4 input modalities simultaneously: text, images (up to 9), video clips (up to 3, ≤15s total), and audio (up to 3, ≤15s total). Mixed input cap is 12 files. Reference-driven generation: you can use an image to lock composition/character appearance, a video clip to specify camera move

Show HN: ZTGI Safety Gateway for LLM Safety

I built a small runtime safety layer for LLM outputs called ZTGI Safety Gateway.This is not a new foundation model and not an AGI claim. It is a post-generation control layer that sits between candidate outputs and final response selection.What it does: - Scores each candidate with two risk tracks: - legacy risk (`p_break`) - hybrid risk (`z_next`: instruction breach + sycophancy + divergence signals) - Enforces hard blocks for: - security abuse prompts - contradiction-actionable prompts

Show HN: MCPlexor – MCP multiplexer that cuts agent context usage by 95%

I built MCPlexor to solve a token waste problem I kept running into with MCP-based agents.The Problem: MCP (Model Context Protocol) is great for giving LLMs access to external tools. But if you connect multiple servers (GitHub, Linear, Postgres, Slack), you end up with 40-50k tokens of tool definitions injected into every request – before the agent even does anything.On a 200k context model, that's 25% gone. On smaller models, it's worse. And most runs only use 1-2 tools.The Solution:

Show HN: Ayder Crash Sandbox – SIGKILL durability proof (per-visitor container)

Hi HN,Most durability demos are “trust me”: a blog post, a benchmark chart, or a unit test.This is a live sandbox that tries to make durability auditable.Every visitor gets:a dedicated Docker container (no shared state with other visitors)a dedicated persistent volume mounted at /dataClick “Run double proof” and it does:Produce N eventsHard-kill the process with a real SIGKILL (docker kill -s KILL)Restart the container (docker start)Consume from offset=0 → should return N (no loss)Commit of

Show HN: Grok Video 10s – Grok AI video generation and creator contest

I’m the creator of https://grok-video.org, a web platform built for high-quality AI video generation.What’s currently available:Grok Video Integration: Full support for Text-to-Video and Image-to-Video using Grok’s video architecture.Fast Rendering: Optimized for speed to help creators iterate quickly.The Roadmap: The current version is powered by Grok, but I am actively working on supporting additional open-weights models like Wan 2.1 and Tencent Hunyuan in the coming updates to provi

Show HN: Sales Agent Benchmark – SWE-Bench for sales AI agents (open source)

Live leaderboard: https://sales-agent-benchmarks.fly.dev/benchmark GitHub: https://github.com/a1j9o94/sales-agent-benchmarkI built an open-source benchmark for evaluating LLMs as sales agents. The idea came from noticing that every sales AI tool demos well on clean summaries but falls apart on real deal data — and there was no rigorous way to measure that gap.How it worksYou register an API endpoint. We send your agent deal context (anonymized real B2B deals),

Why Every Business Must Engage with AI – and How to Do It Right

Title: Why every business should engage with AI (the real question is how deep)AI is no longer an experimental technology. It’s becoming a baseline capability for modern businesses. The real question most teams should be asking is not “should we use AI?” but “how deeply should we engage with it?”I’ve talked to many founders, CTOs, and operators over the past couple of years. The hesitation around AI usually comes from two places:Teams that haven’t really tried AI and feel comfortable sticking wi

5 Best Modems for Xfinity Internet in 2025 – Fast & Reliable

Check the links to all the products with updated prices and details. 1.Arris Cable Modem-https://amzn.to/3VtUzvq 2.Netgear ...