@lennox_saint - Claude Code Setup

Open original

@lennox_saint on X A user on X says you can get better results from Claude Code by picking the Fable 5.2 model with /model and setting sub-agents to Opus 5.5, so they don't copy the main model's output. The post is a short how-to with a prompt template asking for a title, summary, key points, and tags.

  • Steps: open Claude Code, run /model and choose Fable 5.2, then set sub-agents to Opus 5.5.
  • The poster says sub-agents set to the same model copy the main model's output, which is why they recommend the split.
  • The post offers a structured prompt template for big jobs, asking for a highlight, summary, key points, and tags.

The post is a short X thread from an account called lennox_saint, and it is essentially a how-to with a hype opener. The author asks why more people aren't talking about a setup, then lists four steps: open Claude Code, run /model and select "Fable 5.2," set sub-agents to "Opus 5.5" so they don't copy the main model, and give it a large task with a specific summary prompt attached. The core claim is that this configuration produces markedly better results on big jobs, but the post offers no benchmarks, outputs, or comparisons to back that up. That gap is the main thing to know before acting on it. The model names don't match any Anthropic release I can verify, and "Fable" in particular doesn't correspond to a model I'm aware of, so treat the naming as unconfirmed until you check Claude Code's own model list. The sub-agent tip about avoiding copying is also unexplained: the post doesn't say whether it's a documented behavior or the author's observation. Worth a quick test on a small task before committing a large job to it. The one part that's concrete and usable regardless of the model question is the prompt. It asks for exactly three paragraphs, a conversational briefing tone, specifics over vague claims, varied sentence rhythm, and no generic wrap-up. That's a solid template for summarizing any article, and it works equally well with whichever model you're running. Bottom line for your read-it-later queue: the workflow is unverified and the hype outpaces the evidence, but the summary prompt is a useful keeper.

Ask about this article