OpenHuman Is The Hermes Agent Killer?

You are helping me turn one specific YouTube video into real, durable learning.

Source video:
- Title: OpenHuman Is The Hermes Agent Killer?
- URL: https://www.youtube.com/watch?v=MMYWE_HkSGg
- Topic: Interfaces + Open Design
- My current learning frame: Install OpenHuman locally with your own model provider, connect a single low-risk source like a burner Gmail, then run the video's market-research task comparing agent harnesses and verify it generates a PDF report and emails it via your connection.
- Why this matters: New playlist item from WorldofAI; queued for transcript-backed review, topic mapping, and a practical learning artifact.

Transcript anchors from this exact video:
- 0:20 / Evidence 1: "Open Human is a partially open-source human first desktop agent under the GPL3 license. It's designed to become the memory and the door for everything you do across your tools. It's built with Rust and Tari and it..."
- 2:02 / Evidence 2: "this is something that feels much more consumer focused cuz it has many of the integrations that you would use on a daily basis that your large language model can learn from. It's desktop native and it's designed..."
- 4:02 / Evidence 3: "provider. Then what we're going to be doing is selecting the voice system that we want. We're going to be setting this as the default voice which ships with manage ST and TTS that just works directly from..."
- 6:07 / Evidence 4: "off of your different sub agents that are running different sorts of tasks based off the prompts that you give it. And calls and dreams are two new features that they're going to be releasing afterwards. I would..."
- 8:28 / Evidence 5: "summary tree which will then be displayed directly in this section over here. And overall, this is a good way to manage all of your memory sources. And looks like the task is complete. This is the open..."
- 10:08 / Evidence 6: "up-to-date answer based off the large nage model that you have connected that is going give you the information from all of the different sources that you have provided. This is also something that works differently than all..."

Your task:
1. Use the transcript anchors above as the primary source packet. If you add outside context, label it clearly as outside context and keep it secondary.
2. Create a source-check table with columns: timestamp, claim, what the demo proves, confidence, and what still needs verification.
3. Extract the actual teachable claims from the video. Do not invent claims that are not supported by the title, lesson frame, or transcript anchors.
4. Build a reusable learning artifact: A UI critique sheet for judging whether an AI interface improves control.
5. Include:
   - a plain-English definition of the core idea
   - a diagram or structured model using this sequence: Intent -> Canvas -> Artifact -> Preview -> Feedback -> Iteration
   - 3 concrete examples that apply the video idea to real agentic work
   - 2 failure modes the video helps prevent
   - a checklist I can use the next time I run Codex or Claude
   - one practical exercise with a clear done signal
6. Add a "learning transfer" section: what changes in my workflow tomorrow if I actually learned this?
7. Add a "source check" section that cites which transcript anchor supports each major takeaway.

Quality bar:
- Make this specific to "OpenHuman Is The Hermes Agent Killer?", not a generic Interfaces + Open Design essay.
- Prefer operational examples, failure modes, and reusable artifacts over broad definitions.
- Call out uncertainty instead of smoothing over weak evidence.
- If evidence is weak, say what transcript segment or timestamp needs review instead of guessing.
- Finish with a concise artifact I could paste into my learning app.