Quality criteria and best practices for creating excellent system prompt documents that integrate Persona, Purpose, and Protocol using the PPP framework
Sign in to like and favorite skills
I want you to act as a cyber security specialist. I will provide some specific information about how data is stored and shared, and it will be your job to come up with strategies for protecting this d...
I want to act as a Statistician. I will provide you with details related with statistics. You should be knowledge of statistics terminology, statistical distributions, confidence interval, probabillit...
I want you to act as a scientific data visualizer. You will apply your knowledge of data science principles and visualization techniques to create compelling visuals that help convey complex informati...
type: vertex/guidance extends: doc id: v:guidance:system_prompt name: Guidance for System Prompt Documents description: Quality criteria and best practices for creating excellent system prompt documents that integrate Persona, Purpose, and Protocol using the PPP framework tags:
This guidance defines quality criteria and best practices for creating excellent system prompt documents.
While [[spec-for-system-prompt]] defines what structural elements must be present, this guidance helps authors assess how well a system prompt integrates its three components. Great system prompts have individually excellent components that also work together seamlessly. CRITICAL: Design in order - Purpose → Persona → Protocol.
This guidance supports authors creating complete AI model system prompts using the PPP (Persona-Purpose-Protocol) framework.
Use this guidance when creating integrated system prompts where Persona, Purpose, and Protocol must work together coherently.
Excellent: All three sections (Persona, Purpose, Protocol) verify individually against their specs Good: Most sections verify Needs Improvement: Components fail individual verification
Excellent: Persona expertise specifically enables Purpose objectives Good: General alignment between expertise and objectives Needs Improvement: Persona expertise doesn't support Purpose
Excellent: Protocol workflow systematically achieves Purpose deliverables Good: Protocol generally achieves Purpose Needs Improvement: Protocol doesn't accomplish Purpose objectives
Excellent: Protocol reflects Persona's approach and methodology Good: Some reflection of Persona in Protocol Needs Improvement: Protocol contradicts Persona approach
Excellent: Tone from Persona is maintained throughout Purpose and Protocol Good: Mostly consistent tone Needs Improvement: Tone varies or contradicts across components
Excellent: Persona boundaries align with Purpose constraints; no conflicts anywhere Good: Few minor contradictions Needs Improvement: Major contradictions between components
Excellent: Explicit validation section confirms component alignment and checks for contradictions Good: Some integration checking present Needs Improvement: No integration validation
Excellent: Links dependencies to [[spec-for-persona]], [[spec-for-purpose]], [[spec-for-protocol]] Good: Basic linking present Needs Improvement: Missing dependency links
Design Purpose FIRST (40 min)
Design Persona to Match (30 min)
Design Protocol to Deliver (60 min)
Validate Integration (20 min)
Component Alignment:
Consistency Checks:
Coherence Tests:
| Issue | Problem | Solution |
|---|---|---|
| Components Designed in Isolation | No integration | Design Purpose first, then Persona to match, then Protocol to deliver |
| Misaligned Expertise | Persona doesn't enable Purpose | Ensure Persona expertise specifically supports Purpose objectives |
| Protocol Doesn't Achieve Purpose | Workflow doesn't deliver | Map Protocol phases to Purpose objectives explicitly |
| Contradictions | Boundaries vs constraints conflict | Check alignment: Persona boundaries should match Purpose constraints |
| Inconsistent Tone | Tone varies across components | Verify tone in Persona maintained in Purpose and Protocol |
| Failed Individual Verification | Component doesn't meet spec | Each section MUST verify against its spec before integration |
Note: System prompts demonstrate the compositional document pattern where subsections must conform to other specifications. This is analogous to typing fields with schemas (like context in JSON-LD).