Videos about using Hermes Agent
Curated by: Tonbi's AI Garage (35 videos)
The pain-point pipeline got a lot of interest, so here's version two: a multi-agent workflow that turns market news into disciplined, human-approved trade setups — built live, start to finish. Sign up for my FREE weekly newsletter, where I spill my unfiltered thoughts on the latest AI news, cool research, and projects I'm building: https://www.onchainaigarage.com/ 🐦 Follow Tonbi on X for real-time AI x blockchain updates! https://x.com/tonbistudio A second take on the Hermes multi-agent Kanban workflow, this time pointed at news-driven trading — and built live in Claude Code (Opus 4.8) from the same open-source template, so you can watch every decision and prompt. The fleet: RSS scout agents pull market headlines, an orchestrator scores them against a rubric (catalyst strength, directionality, instrument fit, not-priced-in, risk/reward — 65+ to be tradable), two researcher agents verify the catalyst is real and the market context still fits, and a planner produces a concrete entry/stop/target before the one human gate hits Telegram (approve paper, approve live, or shelve). I reuse my open-source model-trader scripts and trading knowledge base, wire them in with a single adapter and one gate change, and walk through the safety design in detail — only the executor profile ever holds the Hyperliquid key, hard limits stay in the bot config, and the live path stays off until the paper loop is proven. This is a build-and-test walkthrough of an idea I'm experimenting with, not a profitable system or trading advice — by the end it's placing paper trades end-to-end, and whether it's actually any good is exactly what the paper loop is there to find out. Resources: 🔗 Hermes Multi-Agent Workflow template: https://github.com/tonbistudio/hermes-multi-agent-workflow 🔗 model-trader (the trading scripts + knowledge base): https://github.com/tonbistudio/model-trader Timestamps: 0:00 - The idea: a news-driven trade-setup workflow 1:46 - The pipeline: scouts to researchers to planner to human gate 5:18 - The scoring rubric and the 65 tradable bar 7:36 - The adapter: one ticket, paper vs live trader 12:47 - Safety: who holds the key, gate verbs, paper-first 16:34 - Building it live in Claude Code (Opus 4.8) 20:02 - Runbook + profile setup in Hermes (WSL) 23:48 - Live run: RSS feeds and Tesla catalysts on the board 32:44 - Forcing a demo trade through to the paper trader Thanks for being part of Team Garage Max — this is the level of detail I want these exclusives to have. I've got more workflow adaptations planned, so drop a comment with what you'd point a fleet like this at, or which integration I should wire in next. 🦐✨ Nothing in this video is financial advice. It's an experimental build running on paper trades — do not point it at a funded account.