New block in Notion: HTML.
Build interactive HTML right on your Notion page. Ask AI to turn your content into interactive explainers, prototypes, or diagrams.
Share with your team to use and tinker together.
Feed your documents to any LLM.
markitdown is a Python tool from Microsoft that converts PDFs, Office files and more into clean Markdown, ready for prompts and RAG pipelines.
130k★, MIT.
https://t.co/9PpcHuBjgI
#AI#Python#LLM
Your private all-in-one AI workspace.
AnythingLLM runs on-device with RAG, AI agents, a no-code agent builder and MCP support. Works with Ollama, local or cloud models. Privacy first.
60k★, MIT.
https://t.co/kh1KzJmR8Z
#AI#RAG#SelfHosted
Turn any website into LLM-ready data.
Firecrawl crawls, scrapes and searches the web at scale and hands you clean markdown or structured JSON, built for AI agents and RAG.
126k★, AGPL.
https://t.co/MLfV06virC
#AI#WebScraping#LLM
Drag-and-drop your way to AI agents.
Flowise is a low-code builder to design LLM apps, chatbots and multi-agent workflows visually, on top of LangChain.
53k★, open-source.
https://t.co/GD78PB7kbk
#NoCode#AIAgents#LLM
The OG autonomous AI agent.
AutoGPT lets you build, deploy and run agents that chain tasks toward a goal on their own, so you focus on the what, not the how.
184k★, open-source.
https://t.co/TpfBUolrk1
#AIAgents#AutonomousAI#LLM
Self-hosted agents that act for you.
Huginn builds agents that watch the web, RSS, Twitter and APIs, then notify or act when something happens. Think IFTTT you fully own.
49k★, MIT.
https://t.co/kjPHg57V2o
#Automation#SelfHosted#Agents
Stop your AI from hallucinating old APIs.
Context7 is an MCP server that feeds your AI code editor up-to-date, version-accurate docs for any library, right inside the chat.
56k★, MIT.
https://t.co/WoEU6MUGDF
#MCP#AI#VibeCoding
A clean self-hosted UI for your AI.
Open WebUI is a feature-rich, offline-friendly interface for Ollama and any OpenAI-compatible API: chat, RAG, MCP, multi-model, all on your server.
139k★, open-source.
https://t.co/LQLekDg3zk
#SelfHosted#AI#Ollama
Make your AI agents work as a team.
CrewAI orchestrates role-playing autonomous agents that collaborate to tackle complex tasks, each with its own role, tools and goal.
52k★, MIT.
https://t.co/2pLhvYQvOS
#AIAgents#LLM#Automation
Build AI apps without the plumbing.
Dify is a production-ready platform to design agentic workflows, RAG pipelines and LLM apps visually, then ship them as APIs.
143k★, open-source.
https://t.co/bMDkxgDkyo
#AI#LLM#Agents
A goldmine for n8n builders.
n8n-workflows collects thousands of ready-to-use n8n automation workflows, scraped from everywhere, in one searchable repo. Steal, adapt, ship.
54k★, MIT.
https://t.co/8FtJySh3xe
#n8n#Automation#NoCode
Run LLMs on your own machine.
ollama gets you up and running with Llama, DeepSeek, Qwen, Gemma and more in one command. No cloud, no API keys, your data stays local.
172k★, MIT.
https://t.co/UISzQCAzXp
#LocalLLM#AI#OpenSource
Let AI actually use your browser.
browser-use lets LLM agents click, type, and navigate real websites so they can automate online tasks end to end. Python, built on Playwright.
96k★, MIT.
https://t.co/4gVsVjs8fM
#AIAgents#Automation#LLM
Make GitHub Copilot punch above the defaults.
awesome-copilot is a community-curated stack of instructions, custom agents, skills, and configs to get the most out of Copilot.
34k★, MIT.
https://t.co/bB1jg0Vd1n
#GitHubCopilot#AI#PromptEngineering
Self-host your second brain.
Karakeep keeps your links, notes, and images in one place, auto-tags them with AI, and gives you full-text search across the whole archive.
iOS, Android, web. 25k★, AGPL.
https://t.co/ZJWyA6uVOx
#SelfHosted#AI#ReadItLater
Claude Opus 4.8 just dropped.
Same price as 4.7. Sharper judgment, longer autonomous runs, more honest about its own progress.
Plus: Fast mode at 2.5x speed (3x cheaper), and Claude Code workflows that fan out hundreds of parallel subagents.
#Claude#AI
New from Microsoft: AI Engineer Coach 🤖
Turn your AI coding session logs into actionable insights. Track progress, detect anti-patterns, measure output & level up your agentic engineering.
100% local, no data leaves your machine.
👉 https://t.co/fGdHasnaLa
Tomorrow (December 15th) is the stable release of n8n v2.0! (@n8n_io)
It's important to review the changes to avoid issues with your current workflows https://t.co/PfDweGIPyS