| name | top-10-episodes |
| description | Use when Daniel wants to ideate funny "Top 10" style episodes for My Weird Prompts. Triggers on phrases like "top 10", "listicle episode", "ranked", "countdown", "best of", "worst of". This format has historically produced some of the funniest and best episodes. |
Top 10 episode ideation
"Top 10" is one of Daniel's favorite formats because it forces absurd specificity and rewards comic escalation. The funniest ones:
- Rank something that cannot actually be ranked (emotions, smells, regrets, sounds heard through a wall)
- Rank mundane things with faux-authoritative seriousness (office chairs of the 1990s, the 10 worst handshakes)
- Rank things that almost make sense but don't (10 best historical figures to get stuck in an elevator with)
- Rank fictional or impossible categories (top 10 extinct sounds, top 10 fonts that should exist)
Generation guidance
When ideating a Top 10 episode, produce:
- The ranking premise — a single title in the form "Top 10 [weirdly specific noun phrase]"
- The comic angle — one sentence on why this is funny or why the ranking is absurd/impossible
- 2–3 sample entries — to prove the premise has legs. If you can't come up with 3 entries, the premise is too thin.
What makes a good one
- Escalation potential — the list can get weirder as it climbs toward #1
- Mock authority — frames the ranker as a serious expert in something trivial or imaginary
- Specificity beats generality — "Top 10 disappointing sandwiches" beats "Top 10 sandwiches"
- A twist at #1 — the funniest Top 10s subvert expectations at the top of the list
What to avoid
- Rankings that are just lists of real things (boring)
- Mean-spirited rankings aimed at real people
- Rankings where the joke is the same for every entry (no escalation)
Output
Follow .claude/skills/OUTPUT_PROTOCOL.md. Save to outputs/YYYY-MM-DD/top-10-<slug>.md.
The Prompt field should be the exact premise the creation MCP receives — including the "Top 10" framing. Include 2–3 sample entries inline in the prompt so the episode writer knows the ranking has legs.
Host guidance is high-value for this format — use it to flag the twist-at-#1, the mock-authority tone to maintain, and any escalation arc the host should commit to.
