Ildar

Drama in one screenshot: 1) Sending Clawdbot email with prompt injection 2) Asking Clawdbot to check e-mail 3) Receiving the private key from the hacked machine ... took 5 minutes That's why we build non-probabilistic agentic security in Archestra: github.com/archestra-ai/a…

Local recommendation system with LightRAG, MCP, Chat UI: archestra.ai/blog/lightrag

Comment 🤑 if you want to get assigned a bounty - $3000 for Agent Browser: github.com/archestra-ai/a… - $900 for MCP UI support: github.com/archestra-ai/a… Bored on long weekends and want to hack something? ☺️ Here are two fun features from our backlog with bounties to try 🚀

Cursor + n8n MCP + Archestra is legitimately game-changing🤯 This MCP integration turns Cursor into a full-stack automation architect. Builds complete workflows with one conversation. Perfect for enterprise teams drowning in manual processes. The problem: Building automation workflows manually is brutal. You're dragging nodes, configuring webhooks, testing connections, debugging failures, all for workflows that break when APIs change. This Cursor MCP setup solves it: → Direct connection to your n8n instance → Create workflows with natural language descriptions → Auto-configures nodes, triggers, and data transformations → Debugs and optimizes workflows in real-time → Generates documentation and monitoring alerts → All built from conversational prompts No node dragging. No webhook wrestling. No broken automations. What you can generate: → CRM sync workflows between platforms → Lead nurturing sequences with multi-channel triggers → Data pipeline automations with error handling → API integrations with custom transformations → Monitoring workflows with smart notifications Built with Cursor MCP. Want the complete setup guide? Comment "N8N" Like this post And I'll send it over (must be following so I can DM)🤯

@DataChaz looks really interesting, this is especially important for mcp server developers, as tool calls blow up context so fast

Cut your LLM costs by 50% 🤯 Stop using JSON → switch to TOON (Token-Oriented Object Notation). It blends YAML’s readability with CSV’s compactness: ↳ 30–60% less tokens ↳ Built-in field validation ↳ GPT-5, Claude, Gemini. Ace for tabular data. Free and open-source 🧵↓

Gateway to manage agents with guardrails

@tom_doerr @tom_doerr Thank you for sharing our project! I would be happy to answer any questions

Today, we have a LIVE demo of Archestra on the official GitHub YouTube channel. We will show how to make n8n agents safe and production-ready with our platform. Join the translation youtube.com/live/SkmluS-xz…

How the Guess Who? game and @simonw's Dual LLM pattern inspired our defense against “lethal trifecta” in our opensource LLM gateway archestra.ai/blog/dual-llm

Grafana OnCall: an easy-to-use on-call management tool grafana.com/blog/2021/11/0…

Really excited about Grafana OnCall, Just announced during the #ObservabilityCON keynote! grafana.com/blog/2021/11/0…

#Amixr is now presenting at #500DigitalDemoDay. Amixr is an open-source incident observability platform for DevOps & SRE to make their life easier. Find out more at amixr.io and tune in here. #b27demoday youtube.com/watch?v=FcfAi3…

Unreal ... 💫 @elonmusk

Vitaly is one of the most active Amixr users and evangelists. A bunch of new features developed based on his feedback to make dxFeed’s incident workflow smooth and effective! blog.amixr.io/case-dxfeed/

@alert_mixer + @HashiCorp = <3 hashicorp.com/integrations/a…

Awesome news from @SlackAPI! @Mkukkk let's update @alert_mixer page

If Microsoft Teams is down for you... Remember it's Office *365* and there are 366 days in 2020.

Incident management without @alert_mixer

Today I first time realised that in QWERTY all the vowels except `a` are on the top row