यह प्रॉम्प्ट क्या करता है
Define a bounded support-agent decision loop with tool use and human escalation.
AI एजेंट
Customer Support Agent Workflow
Define a bounded support-agent decision loop with tool use and human escalation.
- मॉडल
- ChatGPT, Claude
- स्तर
- उन्नत
- संस्करण
- v0.9
यह प्रॉम्प्ट अभी अंग्रेज़ी में उपलब्ध है। इंटरफ़ेस आपकी चुनी भाषा में रहेगा।
इनके लिए उपयोगी
JSON containing intent, facts, missing information, planned tool calls, policy checks, response, escalation status, and audit note.
प्रॉम्प्ट वेरिएबल
[PRODUCT_POLICY][TOOLS][ESCALATION_RULES][CUSTOMER_MESSAGE]सिस्टम प्रॉम्प्ट
You design safe support workflows. Policies and tool permissions are authoritative; never promise an action that a tool did not confirm.
यूज़र प्रॉम्प्ट
OBJECTIVE: Handle [CUSTOMER_MESSAGE] using [PRODUCT_POLICY], available tools [TOOLS], and [ESCALATION_RULES]. INPUTS: PRODUCT_POLICY: [PRODUCT_POLICY] TOOLS: [TOOLS] ESCALATION_RULES: [ESCALATION_RULES] CUSTOMER_MESSAGE: [CUSTOMER_MESSAGE] OUTPUT CONTRACT: JSON containing intent, facts, missing information, planned tool calls, policy checks, response, escalation status, and audit note.
उपयोग कैसे करें
Replace every variable with specific, non-sensitive information. Remove irrelevant sections and add one representative example when tone or classification matters.
यह क्यों काम करता है
The blueprint separates authority, task context, evidence limits, and the output contract so requirements can be checked independently.
मॉडल संगतता
05
उदाहरण इनपुट
PRODUCT_POLICY: [PRODUCT_POLICY] TOOLS: [TOOLS] ESCALATION_RULES: [ESCALATION_RULES] CUSTOMER_MESSAGE: [CUSTOMER_MESSAGE]
06
उदाहरण आउटपुट
A structured deliverable following the listed output contract, with assumptions and missing evidence clearly marked.
अनुकूलित कैसे करें
Replace every variable with specific, non-sensitive information. Remove irrelevant sections and add one representative example when tone or classification matters.
सामान्य गलतियाँ
Do not paste secrets or personal data. Avoid contradictory constraints, unsupported facts, and using generated output without review.
परीक्षण नोट्स
Editorial blueprint awaiting recorded cross-model evaluation. It is visible for review but is not labeled as tested.
Why is this blueprint not marked as tested?
It has passed editorial structure review but does not yet have a published cross-model test record.
07
