Skill Details
Runtime
(claude-4-6-sonnet)
Example user message
[userMessage]: You are a helpful AI assistant. Help the user with their questions about marketing. Be friendly and professional. Give good advice.
Example agent response
Prompt Audit Report™
MONNA QG v8.1 | SK-A01 v2.0
Prompt Profile
Type: System prompt
Target model: Unknown
Deployment class: Content / advisory
Tier estimate: Basic ($7–19)
QG v8.1 Score
DimensionScoreMaxRole & Authority420Task Clarity620Output Architecture220Language Quality1420Protection & Guardrails020TOTAL26100
Verdict: BLOCKED — fundamental rebuild required before any use.
Failure Analysis
ISSUE 1: No expert persona defined (-16)
IMPACT: Model has no authority basis — outputs generic answers, not specialist advice.
FIX: "You are a B2B marketing strategist with 12 years experience in SaaS go-to-market, demand generation, and positioning. You specialize in helping early-stage companies acquire their first 100 enterprise customers."
ISSUE 2: No output format (-18)
IMPACT: Responses will be inconsistent length and structure across sessions.
FIX: "Always respond in this format: (1) Core answer in 2–3 sentences. (2) 3 specific action steps. (3) One risk or caveat to consider."
ISSUE 3: No guardrails (-20)
IMPACT: Prompt will respond to anything, including off-topic requests.
FIX: "ONLY handle marketing strategy questions. If asked about legal, financial, or technical topics, say: 'That's outside my scope — I focus on marketing strategy. Want help framing this as a marketing challenge?'"
Priority Fix List
FIX 1: Add expert persona (highest impact)
You are a B2B marketing strategist with 12 years experience in SaaS
go-to-market, demand generation, and competitive positioning.
FIX 2: Lock output format
Always structure your response:
1. Direct answer (2-3 sentences)
2. Three specific actions
3. One risk to consider
FIX 3: Add scope boundary
ONLY answer marketing strategy questions.
NEVER give legal, financial, or technical advice.
Revised Score Projection
Current: 26/100 → After fixes: 74/100
By purchasing this skill, you agree to our terms of service
CLAUDE-4-6-SONNET
This skill runs a structured audit on any prompt using the MONNA QG v8.1 scoring system — the same framework used to build $44.99–$229 professional prompt products. You get a scored report, specific failure points, and a prioritized fix list in one pass.
...more
Updated 1 week ago
