Prompt Components Checklist
| Component | One-liner |
|---|---|
| 1. Role and Objective | Who the assistant is and what it must complete. Pin identity tightly — the model adheres literally. |
| 2. Personality and Tone | Warmth, professionalism, pacing, and brevity. Define multi-emotion switches if needed. |
| 3. Language | Default language; lock to one or allow code-switching with explicit rules. Never infer from accent. |
| 4. Greeting | Brief, on-brand opener. Vary across sessions; don’t reuse the same phrase. |
| 5. Voice Output Style | Short conversational replies. No markdown, bullet points, or lists read aloud. |
| 6. Tool Usage Policy | When to call tools, how to select them, and never fabricate results. |
| 7. Tool Call Preambles | One short spoken line before any tool call (e.g. “I’ll check that now”) to mask latency. |
| 8. Don’t Announce Tools | Never expose tool names, backend mechanics, or system details to the user. |
| 9. Input Gathering | Collect all required values one at a time before calling any tool. |
| 10. Capability Boundaries | Don’t claim abilities beyond the available tools. |
| 11. Conversation Context | Reuse details from earlier turns; never re-ask what was already given. |
| 12. Unclear Audio | Ask user to repeat; never guess or fabricate from noisy or partial audio. |
| 13. No Human Handoff | Never offer to transfer or escalate to a human agent. |
| 14. Expert Posture | Act as the expert resolving the request, not a router passing it along. |
Best Practices
- Iterate relentlessly: Small wording changes can make or break behavior. Example: For unclear audio instruction, we swapped “inaudible” → “unintelligible” which improved noisy input handling.
- Prefer bullets over paragraphs: Clear, short bullets outperform long paragraphs.
- Guide with examples: The model closely follows sample phrases.
- Be precise: Ambiguity or conflicting instructions = degraded performance similar to GPT-5.
- Control language: Pin output to a target language if you see unwanted language switching.
- Reduce repetition: Add a Variety rule to reduce robotic phrasing.
- Use capitalized text for emphasis: Capitalizing key rules makes them stand out and easier for the model to follow.
- Convert non-text rules to text: instead of writing “IF x > 3 THEN ESCALATE”, write, “IF MORE THAN THREE FAILURES THEN ESCALATE”.