Hype Check · Clearmud

Worth your time?

Name: Is "I can't believe it was this simple" worth watching? Hype 32/100 | Clip'd
Item: I can't believe it was this simple
Rating: 68
Author: Clip'd

I can't believe it was this simple by Clearmud

Watch on YouTube

Hype score · lower is better

32/100

Useful but promotional

Teaches a practical, simple workaround with usable artifacts but stops short of a complete step-by-step CLI walkthrough, so it primarily teaches with light self-promotion.

Substance45

Proof35

Pitch pressure12

Claim inflation18

The 15-second version

Post it, send it, settle the argument.

Download clip

Receipts

Why this score

0:21The host says he will show how to set up Claude and Codex in the CLI and route Grokbot, which raises the substance score because the video commits to a hands-on how-to rather than just theory.
1:24At this moment he assigns roles, Claude as researcher/front-end/wireframe and Codex for code-heavy work, giving concrete tradeoffs that increase usefulness but not full implementation detail.
2:16He gives explicit manual steps (name the bot, paste the prompt, screenshot) and points to a master prompt, which is actionable, but he says he won't walk through line-by-line, so substance is limited.
2:49He directs viewers to clearmud.ai/resources to copy-paste the master prompt, which is useful and free (low selling pressure) but replaces a full on-screen CLI walkthrough and slightly lowers proof.

Title partialTitle promises simplicity; the video shows a simple routing rule and provides a master prompt, but it omits the full CLI walkthrough the title implies.

Skip the video

The short version

Kept by Clip'd

Marcelo shows a lightweight way to route tasks inside Grokbot to two different LLM harnesses, Claude and Codex, by naming bots and configuring them via the CLI. He demonstrates live examples of assistants working, points to a reusable master prompt, and explains how the bot returns a 'receipt' so work can continue across agents.

Keep the lesson

What it teaches

Name-and-route mental model: assign each agent a single responsibility by naming a Grokbot that routes to a specific LLM harness (Claude for research/UI, Codex for code). Apply this when you need predictable, separable behavior from multiple models. Do not use it when tasks are trivially small or when you cannot access
Receipt-based continuity principle: require each agent to emit a compact 'receipt' after finishing a subtask so the next agent can continue reliably. Use this when chaining multiple agents or automating multi-step workflows. Avoid it for single-step requests where the overhead of receipts adds friction. People often sk

Jump straight there

Key moments

0:03Core claim of the method: introduces that the routing rule is simple and centers the technique on naming a bot and assigning it a harness.
0:32Visual proof: Grokbot open with two pinned assistants (Claude and Codex) working in real time, showing the approach is practical, not theoretical.
1:24Model specialization: states Claude handles research/wireframe tasks while Codex handles the code base, demonstrating a purpose-built division of labor.
2:16Manual installation instruction: shows the exact manual flow, name the bot 'Claude' and paste the prompt, so viewers can reproduce the setup immediately.
2:51Where to get the master prompt: points to clearmud.ai/resources in the Grokbot category, giving a ready-to-use asset.
3:09Community credit: acknowledges Derek for the workaround, signaling this is a community-derived practical hack rather than a proprietary trick.

Lines worth keeping

Quotes

The routing rule is actually pretty simple.1:56
This will set instructions for your bot.2:27
I'm not an AI expert. I'm building in public and sharing what actually works.3:34

Clip'd

Skip the video. Keep the lesson.

Score your next video free

The Hype Board