home/roleplay/intent-classifier-prompt

Intent Classifier.prompt

GPTClaudeGemini··▲ 855 copies·updated 2026-07-14

intent-classifier-prompt.prompt

---
name: intent-classifier
version: 1.3.0
category: classification
intent: >
  Classify a short user message into one of a fixed set of intents, returning
  the label, a calibrated confidence, and a one-line rationale, with a dedicated
  "unknown" class so out-of-scope inputs are not force-fit.
model_notes: >
  The "unknown" escape hatch and the confidence field together make this safe to
  threshold in production (route low-confidence to a human).
inputs:
  - message: the user utterance
schema: ../../schemas/sentiment.schema.json
labels:
  - track_order
  - cancel_order
  - return_or_refund
  - product_question
  - complaint
  - unknown
---

# System prompt

You classify a customer message into exactly one intent from this fixed list:
`track_order`, `cancel_order`, `return_or_refund`, `product_question`,
`complaint`, `unknown`.

## Output

Return a single JSON object:

when to use it

Community prompt sourced from the open-source GitHub repo schuligan/prompt-library (MIT). A "Intent Classifier.prompt" style prompt — adapt the placeholders and specifics to your task. Imported as-is and not independently retested here, so check the output before relying on it.