← digests › te9.dev
[ digest / last-10 ]

Last 10 Analyzed.

The 10 most recently analyzed bookmarks from the te9.dev archive. Each entry has been crawled, parsed, and annotated by an LLM for relevance, purpose, and practical use.


gotify/server: A simple server for sending and receiving messages in real-time per WebSocket. (Includes a sleek web-ui)

purpose

Gotify is a self-hosted notification server that allows sending messages through a REST API and receiving them in real-time via WebSocket connections, complete with a web UI for managing users, clients, and applications.

when to use

This resource is most valuable when you need a self-hosted, privacy-focused alternative to commercial push notification services, or when building applications that require real-time event streaming and user notifications without relying on third-party services.

tags
websocket notifications real-time self-hosted rest-api push-notifications messaging docker go

replacements.fyi - performant, safer npm package alternatives

purpose

The site provides a searchable directory of npm packages and suggests modern, performant, native, or more secure alternatives for each. It aims to clean up the JavaScript ecosystem by removing outdated or overly simplistic dependencies.

when to use

It is most valuable during code refactoring, security audits, or when actively trying to reduce a project's overall bundle size and supply chain risks. Developers should also use it before adding a new dependency to see if a native alternative exists.

tags
npm dependency-management performance javascript optimization security bundle-size refactoring

HKUDS/VideoAgent: "VideoAgent: All-in-One Agentic Framework for Video Understanding, Editing, and Remaking"

purpose

when to use

tags

Expose your design system to LLMs

purpose

This article provides a methodology for restructuring existing design systems into LLM-readable formats using spec files, a closed token layer, and automated audit scripts, ensuring AI coding agents produce consistent UI output that adheres to project design standards.

when to use

This resource is most valuable when building prototypes or applications with AI coding assistants over multiple sessions, when maintaining visual consistency across a team using vibe coding, or when integrating an existing design system like Atlaskit into an AI-assisted development workflow.

tags
design systems LLM integration AI coding CSS tokens design consistency vibe coding automated auditing CI/CD spec files token layer

Merlio

purpose

Merlio is an all-in-one AI platform that unifies access to 40+ premium AI models (LLMs, image generators, video creators, voice AI) under one dashboard for a flat monthly fee of $9.99, replacing the need for multiple separate AI subscriptions.

when to use

This resource is most valuable when a developer needs access to multiple AI capabilities simultaneously—such as comparing code suggestions from different LLMs, generating images for web projects, or when budget constraints make individual AI subscriptions ($329+/month combined) prohibitive.

tags
AI platform multi-model AI LLM code assistant image generation productivity AI aggregation SaaS cost savings developer tools

wanshuiyin/Auto-claude-code-research-in-sleep: ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

purpose

ARIS (Auto-Research-In-Sleep) is a collection of plain Markdown skill files that enable autonomous ML research workflows including cross-model review loops, idea discovery, experiment automation, and persistent knowledge management. It works as a standalone CLI or as a plugin within AI coding tools like Claude Code, Cursor, and Trae.

when to use

This resource is most valuable when developers need to automate repetitive research, code review, or experimentation tasks that can run unattended — particularly overnight or during long batch processes. It's also useful when working on ML/AI projects that require iterative paper scoring, experiment tracking, or multi-model collaboration.

tags
AI agent workflow LLM automation Markdown skills ML research Claude Code autonomous agents experiment automation AI coding tools workflow orchestration open source

IvanMurzak/Unity-MCP: AI Skills, MCP Tools, and CLI for Unity Engine. Full AI develop and test loop. Use cli for quick setup. Efficient token usage, advanced tools. Any C# method may be turned into a tool by a single line. Works with Claude Code, Gemini, Copilot, Cursor and any other absolutely for free.

purpose

Unity-MCP is a plugin that connects AI coding assistants (Claude, Cursor, Copilot, etc.) to the Unity Editor via MCP, providing 100+ tools for automating game development tasks like asset management, scene manipulation, and code generation.

when to use

This is most valuable when developing Unity projects and wanting to accelerate development through AI-assisted coding, automated testing, runtime debugging, or when creating custom MCP tools to expose C# methods to AI agents.

tags
Unity MCP AI Game Development CLI Automation Code Generation C# WebGL Plugin

serverless-dna/tendril

purpose

It is an open-source agentic sandbox application that uses AWS Strands Agents SDK to autonomously discover, build, and execute tools on-the-fly when needed, storing capabilities in a persistent registry that grows smarter with each use.

when to use

This resource is most valuable when exploring or implementing autonomous agent architectures, studying agentic tool-use patterns, or building desktop applications that integrate AI agents with persistent, self-extending capability systems.

tags
agentic-ai aws-bedrock tauri strands-agents-sdk tool-building agent-patterns typescript deno-sandbox self-extending-systems desktop-app

yacy/yacy_search_server: Distributed Peer-to-Peer Web Search Engine and Intranet Search Appliance

purpose

YaCy is a distributed peer-to-peer search engine application that includes a web crawler, search index server, and web interface, designed for both public web search and private intranet search deployments.

when to use

This resource is most valuable when you need a privacy-focused, self-hosted search solution for your web application or intranet, when you want to avoid dependency on external search providers, or when you need to index and search private/internal content.

tags
search engine self-hosted peer-to-peer intranet search web crawler privacy open source API Docker Java

HKUDS/OpenPhone: "OpenPhone: Mobile Agentic Foundation Models for AI Phone"

purpose

OpenPhone is a 3B-parameter vision-language model specifically designed for on-device smartphone interaction and task automation, including PhoneClaw, an autonomous AI butler for iOS that can learn user habits and perform complex phone tasks. It provides a complete framework for building mobile AI agents that operate entirely locally without cloud dependencies.

when to use

This resource is most valuable when building mobile applications that require AI capabilities with strict privacy requirements, low latency, or zero API costs, or when developing phone automation features that need to run on-device without internet connectivity.

tags
AI model on-device AI mobile AI open-source vision-language model agentic AI privacy-first edge computing smartphone automation iOS agent