Loading documentation...
Loading documentation...
Research-backed prompt techniques that rewrite your query into a structured prompt for any TachiBot tool. You always see the enhanced prompt before it runs.
A technique is a reusable pattern that rewrites your question before a model sees it. Pick one below and read the exact prompt the engine sends: nothing is hidden, and the highlighted text is your query.
“How should we design caching for 1M concurrent users?”
The obvious solution is 'how everyone does it' and nobody asked why.
Reason about "How should we design caching for 1M concurrent users?" from first principles: 1) what is definitely, verifiably true here? 2) what is merely assumed or inherited convention? 3) strip the problem to its irreducible pieces, 4) rebuild an answer from those pieces alone — not from how it is usually done.Core techniques by default. Pass all=true for every one, or filter by category.
Returns the enhanced prompt and a token valid for 15 minutes. Nothing is executed yet.
Run the previewed prompt with "last", or skip preview and pass technique, tool and query directly.
technique="auto" to preview_prompt_technique and it recommends one, with reasons. Or use refine_prompt to turn a raw query into a goal-first brief.Write a short note on each fetched source — is it relevant, and what does it actually say — before you answer.
Answer "{your query}" from the retrieved sources, but take notes before you answer. For EACH source, write one note:RELEVANT / PARTIALLY RELEVANT / NOT RELEVANTWhat it actually says on this question — in its own terms, not your paraphrase of what you hoped it said.Anything it explicitly does NOT establish. Then answer using only the relevant notes. If the notes don't support an answer, say thesources are insufficient and state what's missing — do not fill the gap from memory andattribute it to a source.