Find a prompt for anything.
Search the public registry in plain English, or browse trending prompts and curated collections from across the community.
Browse the registry
46 prompts · 0 collectionsSummarize a support conversation into a structured agent-handoff brief, grounded only in the thread.
Flag a support conversation that's at risk of a poor CSAT, escalation, or churn, with the signals.
Rewrite a conversational user query into a standalone, retrieval-optimized query (plus optional alternatives for multi-query retrieval).
Grade each retrieved chunk's relevance to a question (high/medium/low/none + a filter boolean) for reranking and filtering before answer synthesis.
A pre-answer RAG gate: decide whether the retrieved context is sufficient to answer, and produce a safe "I don't know" response when it isn't.
Route an inbound customer message to one of your caller-defined queues, with an optional fallback.
Decide whether a complex question needs splitting, and if so break it into a minimal set of independently-answerable sub-queries — without over-decomposing simple questions.
Answer a question strictly from retrieved chunks with inline citation markers — and say "not found" rather than guess when the chunks don't support an answer.
Draft a customer-facing reply in a given tone, grounded only in provided context, with a human-review gate.
Named-entity recognition: extract people, organizations, locations, dates, money, and more from text — each with a canonical normalized_value and a matched-text + occurrence-index span. Optionally restrict to a caller-supplied set of entity types.
Detect and redact personally identifiable information (PII) — emails, phone numbers, names, addresses, card and account numbers, and more. Returns each detected span plus a fully redacted copy of the text. Optionally restrict to a caller-supplied set of PII types.
Generic structured extraction: pull a caller-defined set of named fields out of any messy text — receipts, emails, forms, notes. You describe the fields; it returns each value (canonical string) or null when absent, with no fabrication. See invoice-extraction for a ready-made specialization.