Show HN: Pylon Sync, an agent-first full-stack realtime framework https://ift.tt/8NAPYi2

Show HN: Pylon Sync, an agent-first full-stack realtime framework I created Pylon to make it easier to move from hobby projects to full production apps. When I work on hobby projects, I usually use React or Next.js because they are quick to set up and easy to deploy on Vercel. For production apps, I separate the frontend and backend, then deploy the backend on AWS. But setting up a full backend on AWS can be complex and costly, especially for simple apps. Pylon is a full-stack, real-time framework that includes server-rendered React, TypeScript functions, entities, policies, real-time sync, built-in authentication, and support for background and scheduled jobs. By default, it uses SQLite, but you can switch to Postgres for production. The authentication system is heavily inspired by better-auth. The runtime is a Rust server that runs TypeScript functions and server-rendered React using Bun. Pylon itself is inspired by Rails and focuses on convention over configuration, so you have fewer decisions to make before deploying. This approach applies to modern React apps, real-time sync, TypeScript server functions, authentication, job management, and deployment. One of Pylon’s main goals is agent compatibility. It lets coding agents build and deploy apps with no setup, quick understanding, secure defaults, and easy deployment, all without requiring any third-party services. Pylon works for both quick projects and production apps where performance, observability, ownership, and self-hosting matter. While it’s easy to self-host Pylon apps, Pylon Cloud provides managed hosting with a developer experience similar to Vercel. You can deploy from git or the CLI, get an instant URL, add custom domains, and go live in seconds. Each app runs on its own server, which can scale to zero, with TLS and global caching enabled. If you have experience with Next.js, Vercel, Convex, Supabase, Firebase, better-auth, or Rails, I’d love to hear your feedback. Create your first app: npm create @pylonsync/pylon@latest Website: https://ift.tt/ASyXz4N Repo: https://ift.tt/wRbXkjs Docs: https://ift.tt/vU0KrC4 LLMS: https://ift.tt/xuzi65y Skill: npx skills add pylonsync/pylon Examples: https://ift.tt/IuLUKB1 https://ift.tt/ASyXz4N July 9, 2026 at 09:38PM

Show HN: Policy enforcement for Claude Code, Cursor, and Codex https://ift.tt/GyOT9sF

Show HN: Policy enforcement for Claude Code, Cursor, and Codex Show HN: Runtime authorization for Claude Code, Cursor, and Codex Hi HN, Fernando and I built Kastra. Kastra intercepts AI agent tool calls and evaluates them against deterministic policies before they execute. This is aimed at developers using coding agents like Claude Code, Codex, Cursor, and OpenClaw. We built Kastra after one of our Cursor agents almost executed DELETE FROM customers WHERE status='test' against a production database. We caught it before it ran, but it made us realize that nothing in our stack actually decided what the agent was allowed to do. What mattered for us wasn't the mistake; it was realizing nothing in our setup would have stopped it if we weren't actively on top of it. LLMs are probabilistic, and prompts influence behavior, but they don't deterministically decide what an agent is allowed to do. Without a deterministic policy system, nothing could have decided what it was allowed to do. Kastra pushes an allow, hold, and deny decision before the action runs. You can build these policies in plain English from the web app. The interception engine evaluates the tools, targets, and parameters of every action. We also shipped many policy packs covering common high-risk scenarios, and every decision is recorded in an immutable audit trail. The desktop app, CLI, dashboard, and Recon scan are free to use for developers. If you often use Claude, Codex, Openclaw, and Cursor, Kastra can run a scan command on which risky actions your agents have already taken and automatically build rules to avoid them from happening again. Recon is a feature of Kastra that scans your local agent history. In order to run this scan, execute the commands below in your coding agent. brew install kastra-labs/tap/kastra-edge kastra-edge scan The scan reads your local agent session history, and it shows all the risky actions your agent has already taken before, the secrets written to tracked files, production databases touched, force pushes, curl-to-shell, and more. This runs on your machine, and secrets never leave. In our own use cases, we kept finding things we'd forgotten or didnt know agents had done. Each finding can be converted into a runtime policy, letting you delegate more work to AI without trusting the model itself. Kastra intercepts all workloads at runtime and makes sure these policy evaluations typically complete in under a millisecond. Instead of trusting the model, you trust the deterministic rules that govern its actions. One problem we are still working on to improve the stack is how to manage teams of agents with conflicting policies. We would love feedback from anyone building multi-agent systems. Fernando and I will be reviewing the comments. We are super curious what your first scan finds. Please post results below so we can see what the most common patterns are and adjust policy packs for our users based on your feedback. Documentation: https://kastra.ai/docs Download for MacOS Kastra Edge: https://ift.tt/6fq8Fge Check Kastra in action today: https://www.youtube.com/watch?v=6TUETu5lb3Q&feature=youtu.be https://kastra.ai/ July 9, 2026 at 07:26PM

Show HN: Getting GLM 5.2 running on my slow computer https://ift.tt/QtcO9f7

Show HN: Getting GLM 5.2 running on my slow computer A few days ago I found myself trying out GLM 5.2 and was really positively impressed. The capabilities and security I was getting from this LLM are similar to those I've gotten from models like Claude or GPT, and this really surprised me. But then I thought, "I wonder how it would work on a normal computer like mine," and above all, "I wonder if it would work without going into OOM on a computer like mine." So I started working with the help of agents to test this possibility. I started converting the model to int4, understanding MTP usage, and if possible implementing DSA for long context. How it responds in int4 and whether the quality is maintained or not. Until I got to the point, on my computer with 32GB of RAM, I was able to communicate with GLM 5.2 with times that, of course, aren't high in cold start, but even then, we're talking about 0.1 tok/s, but that wasn't important to me. The important thing was the journey to reach this goal. I just wanted it to work at all costs, even slowly. So I created Colibrì, which was born from a very simple idea, to be honest, but tested in every way, where a 744B Mixture-of-Experts model activates only ~40B parameters per token—and only ~11 GB of those change from token to token (the routed experts). So: The dense part (attention, shared experts, embeddings—~17B params) stays resident in RAM at int4 (~9.9 GB); The 21,504 routed experts (75 MoE layers × 256 experts + the MTP head, ~19 MB each at int4) live on disk (~370 GB) and are streamed on demand, with a per-layer LRU cache, an optional pinned hot-store, and the OS page cache as a free L2. The engine is a single C file (c/glm.c, ~1,300 lines) plus small headers. No BLAS, no Python at runtime, no GPU.No GPU or serious hardware because I don't have that hardware so I can't test it on hardware that is more powerful than my computer.Colibrì is a one-person project, written and tested entirely on a 12-core laptop with 25 GB of RAM — the numbers above are the ceiling of what I can measure at home. Any feedback is welcome! (and if anyone wanted to participate in the project I would be delighted) Repo: https://ift.tt/PMSBqRx https://ift.tt/PMSBqRx July 9, 2026 at 12:05PM

Show HN: EVconomics – EV vs. gas cost-of-ownership calculator with live prices https://ift.tt/C7NVZs6

Show HN: EVconomics – EV vs. gas cost-of-ownership calculator with live prices https://ift.tt/EGkHATl July 9, 2026 at 11:29PM

Show HN: Onboard-CLI, a LLM powered and AST-based tool to visualize codebase https://ift.tt/Ap03trd

Show HN: Onboard-CLI, a LLM powered and AST-based tool to visualize codebase https://ift.tt/I3JzxYt July 9, 2026 at 12:09AM

Show HN: Skill-extractor turns coding agent transcripts into reusable skills https://ift.tt/rjNzpvD

Show HN: Skill-extractor turns coding agent transcripts into reusable skills https://ift.tt/czvlwK1 July 9, 2026 at 12:03AM

Show HN: Hnwork.app – UI for Who is hiring posts https://ift.tt/lVYsoOL

Show HN: Hnwork.app – UI for Who is hiring posts Hey HN, I built a UI on top of the "Who is hiring" posts. Take a look at https://hnwork.app ! One of the downsides of unstructured text posts is the readability due to it being free-form and having little to no format. While there are other tools that have been built over the years to make perusing Who is hiring posts easier, I took a try on making my own (I actually tried to build this at a YC hackathon a few years back, but got around to completing it recently). Features: - Text search and search filters - Original post text with call outs to important information - Removes posts that aren’t on topic (complaints, seeking work, vague or missing contact info) - Analytics - API In addition, job posters can create accounts to submit postings through the app. While I don’t expect posting to move over to this app, it’s what I envisioned what a Who is hiring thread would like as an app: - Structured postings with required fields (e.g., salary range required) - Job posters get notifications about comments on their posts - Job posters get verified through their email before posting (e.g., someone posting a Sony job has a Sony email address) - Companies with multiple job posters can coordinate postings and view past postings - Admins can audit and approve companies and posts Job seekers can also create an account to post comments or get access to a simple API but otherwise browsing doesn’t require any kind of signup/signin. I’m open to feedback: let me know if you’d like me to ingest more data from past months, something is missing or broken, or there’s a new feature you’d like to see. Thanks! https://hnwork.app/ July 9, 2026 at 12:00AM