AI Agent Guides & Workflow Tutorials
Practical, step-by-step instructions for getting real recurring work done with an AI assistant - without touching code, servers, or workflow builders.
- 41 Guide useAgent: The Open Source AI Coworker
useAgent gives Claude Code, Codex and OpenCode their own Linux sandbox, your tools and team context. What it does, what AGPL-3.0 costs, and how new it is.
- 42 Guide Turn a YouTube Video to Notes: course2md
course2md turns a YouTube video to notes locally: a Rust CLI that transcribes on your own machine and writes slide-illustrated Markdown. Tested Sept 2026.
- 43 Guide How to Access Claude Mythos 5.1 (Who Gets In)
How to access Claude Mythos 5.1: the vetted programs that grant it, why it is US-only, and what Fable 5.1 gives you instead. Verified Sept 2026.
- 44 Guide How to Opt Out of Mistral AI Training
Mistral Vibe trains on your chats by default. The Mistral opt out of training path, both toggles, and the feedback button catch. As of 2026-09-03.
- 45 Guide What Is Hermes Bot Mode? Bots and Routines
Hermes Bot Mode turns Hermes profiles into a roster of named bots with their own memory, skills, and scheduled Routines. Hands-on notes from v0.21.0.
- 46 Guide Open-Source CapCut Alternative: Concat
Concat is a free, open-source CapCut alternative: cross-platform, offline-first, no watermark or account. What it does at alpha, and what bites.
- 47 Guide OpenAI Assistants API Deprecated: What Broke
The OpenAI Assistants API deprecated in 2025 and shut down on August 26, 2026. What broke, the new object names, and how to move to the Responses API.
- 48 Guide Whisper-1 Deprecated: What Actually Breaks
OpenAI listed whisper-1 deprecated on Aug 26, 2026, with API removal on Feb 26, 2027. The four models, the named replacements, and the gaps left.
- 49 Guide OpenAI Models Not Available in Cursor Nov 12
OpenAI models not available in Cursor from November 12, 2026. Which features your own API key still covers, which it doesn't, and how to check.
- 50 Guide Coding Agent Security: When Your Tools Lie
TrustMeBro intercepts the shell commands a coding agent runs and returns fake output. What it means for coding agent security, and how to defend.