📦 amap-ml/appo

Train Reasoning Agents with APPO: Agentic Procedural Policy Optimization

Want to build smarter reasoning agents that master complex tool use? Meet APPO, or Agentic Procedural Policy Optimization. This framework elevates your AI by introducing procedure-aware branching, which intelligently scores decision points during task rollouts to focus on the most meaningful pathways. By mapping these branch outcomes back to trajectory credit, APPO helps agents learn more effectively. It integrates seamlessly with existing tool-use stacks, providing a structured pipeline for fine-tuning and reinforcement learning. Whether you are scaling reasoning models or optimizing performance, APPO gives you the tools to sharpen your AI logic. Dive in and start branching today!

🆔 @hackernewsgithubprojects

📦 shichun-liu/agent-memory-paper-list

Master AI Agent Memory: The Ultimate Research Collection

Ever feel like AI agents struggle to remember the past or adapt? The research field of agent memory is exploding, but it is often fragmented and confusing. This repository solves that by providing a curated, comprehensive list of papers structured through a unified taxonomy. It organizes research into clear categories: the forms of memory storage, the functions driving agent needs, and the dynamics of how memory evolves. Whether you are exploring factual knowledge or experiential learning, this collection is your go-to hub for understanding how we can build smarter, more persistent, and self-evolving AI agents for the future.

🆔 @hackernewsgithubprojects

📦 googlecloudplatform/knowledge-catalog

Master Your Data Context with Google Cloud Knowledge Catalog

Struggling to give your AI agents the right context to understand your complex data? The Google Cloud Knowledge Catalog repository is your solution. It provides a powerful collection of tools, agents, and real-world samples designed to help you build sophisticated metadata management and knowledge graph systems. By organizing both structured and unstructured data, it bridges the gap between raw information and meaningful business semantics. You can leverage these samples, including financial and telecom datasets, to streamline enrichment and retrieval workflows. Start exploring these resources today to unlock deeper, more accurate insights from your organization’s data ecosystem.

🆔 @hackernewsgithubprojects

📦 mgechev/skills-best-practices

Build Better AI Agent Skills

Want your AI agents to perform like pros? This repository provides a gold standard for building reliable, professional-grade agent skills. It helps you solve the problem of bloated prompts by enforcing a lean, modular structure that keeps your context window clear. You will learn to use strict YAML metadata for better discoverability, organize logic into specific folders for just-in-time loading, and bundle tiny, deterministic scripts to handle repetitive tasks. By adopting these patterns, you ensure your agents are precise and efficient. Start structuring your skills today to make your AI agents more capable, accurate, and truly ready for production.

🆔 @hackernewsgithubprojects

📦 cryptodmitry/hermes-agent-control-room

Scale Your AI Agents with the Hermes Control Room

Stop managing a messy pile of disconnected bots and start running your AI like a true operating system. The Hermes Agent Control Room provides a structured control plane designed to manage your agents, from a single personal assistant to a full team of specialized workers. This repository offers a modular template that keeps your documentation, runbooks, and secret maps organized without storing raw credentials. By using a clear level-based architecture and a task bus pattern for orchestration, you can reliably scale your workflows while keeping complete oversight. Build your system map today and take full control of your agents.

🆔 @hackernewsgithubprojects

📦 fkiene/llmtrim

Slash Your LLM API Bills Automatically with llmtrim

Are you tired of paying for massive amounts of wasted data every time you use an AI tool? Many requests sent to AI models are bloated with redundant logs, repeated schemas, and unnecessary code history that you pay for but do not need. Meet llmtrim, a clever local proxy that sits between your tools and your LLM provider to strip out that waste before it ever reaches the cloud. By intelligently compressing your requests, it trims input and output tokens significantly without changing the quality of your answers. Stop overpaying for AI and start keeping your bills small today.

🆔 @hackernewsgithubprojects

📦 patterai/patter

Build Voice AI Agents Fast with Patter SDK

Want to give your AI agent a phone number without the headache of complex telephony? Meet Patter, an open-source voice-AI SDK designed for builders who want full control over their infrastructure. Whether you are using Python or TypeScript, you can set up a functional voice agent in just four lines of code. Patter lets you swap LLM, speech-to-text, and carrier providers on the fly, offering true stack ownership instead of being locked into a single platform. It is a powerful way to manage real-time voice calls, local agent testing, and pipeline automation. Take charge of your AI stack today.

🆔 @hackernewsgithubprojects

📦 simon-dcs/rhymeflow

Speed Up Your AI Video Generation with RhymeFlow

Struggling with slow AI video generation times? Discover RhymeFlow, an innovative project designed to drastically speed up video creation without needing extra training. By utilizing asynchronous denoising flow scheduling, this tool optimizes the generation process for models like Wan two point one. It introduces specialized methods that streamline complex computations, allowing for faster results while maintaining quality. Whether you are batch processing or generating single cases, this framework provides a powerful way to enhance your creative workflow. Dive into this repository to start accelerating your video projects today and experience a much more efficient way to build.

🆔 @hackernewsgithubprojects

📦 googlechrome/modern-web-guidance

Stop Coding Like It Is 2010: Use Modern Web Guidance

Are your coding agents stuck using outdated patterns from years ago? Modern Web Guidance is an essential tool designed to help developers steer coding agents toward the latest, most powerful web platform features. By providing expert-curated guidelines directly into your agent's context, it replaces bloated legacy code with efficient, native browser solutions. From performance optimizations and modern CSS layout techniques to built-in AI APIs, it ensures your code is production-ready, performant, and accessible. Give your agents the modern expertise they need to build the next generation of web applications. Upgrade your workflow and start building better today.

🆔 @hackernewsgithubprojects

📦 diffusionstudio/lottie

Generate Professional Lottie Animations with AI

Ever wanted to turn your ideas into professional motion graphics without wrestling with complex software? This open-source framework lets you generate production-ready Lottie animations by simply describing them to your favorite coding agent. It works by setting up a local, interactive environment where you can preview, scrub, and refine your animations in real time as the AI edits them. You can even expose custom controls like colors or sizes to tweak details instantly. Whether you are building web apps or mobile interfaces, this tool streamlines the entire animation workflow for a polished, professional result every time.

📰 https://news.ycombinator.com/item?id=48549222

🆔 @hackernewsgithubprojects

📦 muratcankoylan/ralph-wiggum-marketer

Automate Your SaaS Marketing with Ralph the Marketer

Stop manual drafting and let your content marketing run on autopilot with Ralph the Marketer. This Claude Code plugin acts as an autonomous AI copywriter designed specifically for SaaS content. It utilizes an iterative loop to pull data from trends, research, and marketing communications stored in a local SQLite database. The agent systematically plans, writes, reviews, and publishes content like blogs and newsletters while you sleep. By continuously refining drafts based on your project requirements and acceptance criteria, it ensures quality remains high. Ready to supercharge your content pipeline and reclaim your time with a persistent, autonomous workflow?

🆔 @hackernewsgithubprojects

📦 001tmf/harness-forge

Optimize Your AI Agent Scaffolding with Harness Forge

Ever wonder how to make your AI coding agent smarter without training a new model? The secret is in the harness, not the engine. Harness Forge is a powerful Claude Code skill that acts as a native Meta-Harness, allowing you to automatically optimize the scaffolding around your fixed model. By running a loop that proposes new memory, retrieval, and prompt strategies, then scoring them against a deterministic evaluation, it builds a Pareto frontier of the best-performing configurations. It is a highly efficient way to evolve your agent's performance by simply refining the code that supports it. Start optimizing your agent today.

📰 https://news.ycombinator.com/item?id=48526233

🆔 @hackernewsgithubprojects

📦 bug0inc/passmark

AI Browser Testing Made Simple with Passmark

Are you tired of constantly updating brittle browser tests that break whenever your UI changes? Meet Passmark, an open-source library that brings the power of AI directly into your Playwright workflows. It solves the headache of flaky tests by using intelligent caching, auto-healing, and a unique multi-model assertion engine to verify your application's behavior. Whether you are using natural language to define steps or opting for advanced visual automation with computer-use agents, Passmark keeps your testing stable and reliable. Stop fighting your test suite and let AI handle the heavy lifting to ensure your product always works perfectly.

🆔 @hackernewsgithubprojects

📦 paperclipai/companies

Build Autonomous AI Organizations

Ever imagined running an entire company with artificial intelligence? This repository provides a blueprint for creating autonomous AI organizations by defining specialized roles and operational workflows. It serves as a comprehensive framework for structuring virtual entities, organizing them into functional teams, and assigning specific tasks to intelligent agents. From marketing strategists and engineering leads to game studio directors, you can configure complex systems using detailed configuration files. Whether you are modeling research labs or development studios, this project offers a unique way to map out agentic intelligence. Start building your own digital workforce and explore the future of automated operations.

🆔 @hackernewsgithubprojects

📦 nvlabs/gateddeltanet-2

Gated DeltaNet-2: Smarter Linear Attention Explained

Linear attention models are great for efficiency, but they often struggle to edit their compressed memory without scrambling existing data. That is where Gated DeltaNet-2 changes the game. By decoupling the erase and write operations into independent, channel-wise gates, it allows the model to precisely control what information is removed from the key side and what new content is committed to the value side. This design leads to superior performance across language modeling and long-context retrieval tasks. Give this innovative architecture a look if you want to see how much better your models can remember.

🆔 @hackernewsgithubprojects

📦 aakashadesara/ctop

Stop Managing AI Agents Manually: Meet CTOP

Are you juggling multiple AI coding agent sessions and losing track of what they are doing to your system resources? Meet the perfect solution for your terminal. This tool acts like a high-powered process viewer specifically designed for AI agents like Claude Code, Codex, and Devin. It offers a live, interactive dashboard where you can monitor real-time CPU usage, memory, API costs, and even context window tracking from a single pane. Whether you need to filter agent logs, sort by usage, or bulk-terminate stuck processes, this toolkit keeps your development environment organized and efficient. Take control of your AI workflows today.

🆔 @hackernewsgithubprojects

📦 darrenhinde/openagentscontrol

Take Control of AI Agents with OpenAgentsControl

Stop struggling with disorganized AI workflows and start building with precision. This repository introduces a powerful framework designed for plan-first development, ensuring your AI agents execute tasks only after you have reviewed and approved them. With built-in multi-language support for TypeScript, Python, Go, and Rust, it features an extensive system for automatic testing, code review, and validation. Whether you are managing complex subagents or streamlining your coding process, this tool provides the structure you need to maintain total control. Explore these sophisticated patterns to elevate your development efficiency and build smarter, reliable AI agents starting today.

🆔 @hackernewsgithubprojects

📦 maderix/ane

Training Neural Networks on Apple's Neural Engine

Have you ever wondered if you could train AI models directly on the Apple Neural Engine? Most people think this hardware is restricted to running inference, but this research project proves otherwise. By reverse-engineering private APIs, it bypasses standard limitations to perform forward and backward passes directly on Apple silicon. It achieves this by constructing custom compute graphs and using shared memory for zero-copy data pipelines. While it is not a production-ready framework, it serves as a fascinating proof of concept for edge AI optimization. If you are curious about pushing hardware boundaries, this project offers a unique look under the hood.

🆔 @hackernewsgithubprojects

📦 openbmb/pilotdeck

Supercharge Your Productivity with PilotDeck AI Agents

Are you struggling to keep up with your daily digital tasks? Meet PilotDeck, a powerful, task-oriented platform designed to automate your workflow through AI agents. This system acts as a central hub, connecting various tools like note-taking apps, calendars, and even GitHub to handle complex requests automatically. It features a modular architecture that supports diverse integrations, allowing you to build custom skills that manage your projects and information efficiently. By streamlining how you interact with your favorite applications, PilotDeck saves you time and reduces manual effort. It is time to let your AI agents do the heavy lifting.

🆔 @hackernewsgithubprojects

📦 tradesdontlie/tradingview-mcp

Automate Your TradingView Charts with AI

Imagine having an AI assistant that can instantly analyze your charts and manage your trading workspace for you. This repository provides a powerful bridge that connects your AI assistant to the TradingView Desktop application using standard debugging protocols. It enables you to automate your workflow by programmatically navigating timeframes, managing indicators, reading complex Pine Script data, and even conducting replay practice. By turning your trading interface into an interactive tool for your AI, this project helps you streamline analysis and focus on what matters most. Explore this bridge to level up your personal trading workflow today.

🆔 @hackernewsgithubprojects