Product Radar
A unified feed of emerging software, developer tools, and early-stage market launches.
Domphy, a UI framework.
A simpler, clearer UI framework alternative to React, using plain JS objects and functions for state, designed for human readability and AI-correctable output. Specifically used for SketchUp and Revit plugins in AEC.
BlitzGraph, a graph database designed as a "Supabase for graphs" and specifically built for LLM agents, featuring polymorphic records/relations, GraphQL-like queries, and a JSON query language.
Positioned as a next-generation graph database that overcomes the limitations of SQL and existing graph databases (Dgraph, Typedb, SurrealDB, Neo4j) by offering a more flexible data model, agent-centric design, and integrated app development features. It aims to "fully ditch t...
NAADI
Corporation tax automated. So accountants advise, not admin
Ferrix AI, an agentic platform designed to automate product management workflows, specifically research synthesis, spec writing, and progress tracking.
Positioned as a solution to the "bottleneck" in product decision-making, which has emerged as AI agents accelerate engineering. It aims to make product managers more efficient by automating mechanical tasks, mirroring how AI assists engineers.
git-lrc – Free, Micro AI Code Reviews That Run on Git Commit
A tool for individual developers to perform quick, self-directed AI-assisted code reviews at commit time, focusing on common risk patterns across various categories (security, reliability, performance, maintainability). It aims to instill a habit of understanding and vouching ...
Harness Starter Kit
Repo guardrails for reliable AI coding agents
Steal-a-GIF – Browser tool to export GIFs from locked-down platforms.
A tool to turn a screen recording into a slack-sized emoji, specifically for platforms like TikTok and WeChat that prevent saving/downloading.
Building a startup's full site using Claude Mythos (an LLM).
A demonstration of extreme rapid prototyping and development, claiming a full site build in one day using an advanced LLM (Claude Mythos), highlighting the speed benefits of AI-driven development.
Outbound Rewriter that gets replies
Paste your pitch, get 5 scored variants + follow-up cadence
Claumon – forecasting Claude Code usage limits with a Gamma process
An all-round control panel for Claude Code, offering usage forecasting with a calibrated statistical model (Gamma process), consumption gauges, cost breakdowns, conversation history, and memory management, running locally as a single binary.
Foyer – Learn while you wait for your agents to code
A tool designed to help developers learn about what their AI agents (e.g., Claude Code) are doing during wait times, preventing loss of focus.
A web-based tool that provides precise, localized information for Red Flag Warning zones (specifically NWS zone CAZ515 for East Bay Hills). It offers a yes/no check for address inclusion, Genasys evacuation zone, real-time wind/humidity data, a plain-English action checklist, a per-school decision view, and a one-tap iMessage buddy-check template. No app installation is required.
Solves the problem of inaccurate, county-wide emergency alerts by providing precise, localized information about Red Flag Warning zones and associated risks. Positions itself as a critical, rapid-response tool for public safety, enhancing individual preparedness.
CapaKit, a sandbox for the entire AI-app lifecycle (build, test, run).
Makes agent-driven development safe and productive by sandboxing the entire app lifecycle, not just runtime, addressing security gaps in AI agent deployment.
Ccgs (Collaborative Claude Code sessions), a tool to store and share Claude Code sessions using Git branches.
Solves the problem of Claude Code sessions being 'trapped' locally and unshareable, enabling colleagues to share and resume AI-assisted development sessions via Git orphan branches.
Chatcode, a universal layer for persisted agentic sessions, interacting via web and messenger (Telegram), for Claude Code and Codex.
Remote Control for Claude Code and Codex. A 'universal layer for persisted agentic sessions.' Free to start.
MetaBrain, a local document memory for AI agents, with human collaboration features.
Addresses the need for AI agents to track contextual data beyond 1D chat and dynamically retrieve specific knowledge. Also allows human collaboration.
Lite-Harness, a self-hosted Dockerfile providing a simple harness server for running AI agents with built-in memory, durable sessions, cron scheduling, and a vault.
A simple, self-hosted harness server to run AI agents, offering out-of-the-box memory, durable sessions, cron scheduling, and a vault.
An interactive, in-browser 3D stellar map allowing exploration of local star systems, interstellar travel planning, exoplanet viewing, stellar characteristics, and 3D constellations, rendering 600k stars from GAIA DR3 mission data.
Local Stellar Map
A tool to convert LibreOffice .odt files into publishable file bundles for Kindle.
Self Publish Studio
Multiplayer, a local debugging agent that runs alongside coding agents (e.g., Claude Code, Codex, Copilot) to capture full-stack, unsampled session data (frontend actions, backend traces/logs, request/response content/headers) only when issues occur, then deduplicates them before feeding to the coding agent.
Solves the problem of 'PR slop' caused by coding agents inheriting limitations from existing observability stacks (sampled traces, aggregated metrics, limited context). Positions itself as providing a 'complete, correlated picture of what actually broke' by capturing unsampled...
SaaS Metrics
Hacker News
Product Hunt