Finally, a Programmable AI Agent Framework That Works

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

Source video:
- Title: Finally, a Programmable AI Agent Framework That Works
- URL: https://www.youtube.com/watch?v=n5cYS6KuyK8
- Topic: Codex + Claude Workflows
- My current learning frame: Build a Flu YouTube-title workflow that loads a script file as a prompt and uses a scoring skill, first running it in the default just-bash sandbox to see it fail, then making it work via either local execution or a custom tool.
- Why this matters: New playlist item from Better Stack; queued for transcript-backed review, topic mapping, and a practical learning artifact.

Transcript anchors from this exact video:
- 0:00 / Evidence 1: "This is Flu, an open-source framework for building AI agents made by the team behind Astro, which basically takes what Claude Code does as a harness and makes it 100% programmable. So, things like skills, tools, sandboxes are..."
- 1:34 / Evidence 2: "that same limitation. With Flu, you can build agents like Claude Code that require human input, but you can also build workflows that don't require a human at all, which is useful for agentic processes that are very..."
- 3:09 / Evidence 3: "system prompt. And that's it. That's all you need to create a basic Flu agent. Let's see it in action. To run it, we can use Flu connect, which builds and runs the agent, and we're matching the..."
- 5:01 / Evidence 4: "rank them. So, let's see if this works. But before we do, I want to make it obvious that workflows are in the workflows directory and skills are in the skills directory from the project's root. And to..."
- 7:23 / Evidence 5: "summary of the results. Of course, Flu supports WebSockets instead of HTTP if you wanted to stream the workflow information. Now, as usual, there are so many features Flu has that I didn't get round to going through,..."

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 routing matrix for when to use Codex, Claude, browser checks, or manual review.
5. Include:
   - a plain-English definition of the core idea
   - a diagram or structured model using this sequence: Inspect -> Plan -> Edit -> Verify -> Review -> Route
   - 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 "Finally, a Programmable AI Agent Framework That Works", not a generic Codex + Claude Workflows 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.