FAI Boost Prompt

Scores prompts on clarity, specificity, and constraint coverage. Recommends improvements: add role/context, use few-shot examples, define output format, add guardrails. Python tool analyzes prompt patterns; suggests refinements; A/B tests against baseline. Integrates with Prompt Flow evaluation pipelines.

Workflow

Step 1 — Analyze Prompt Quality

from fai_boost import PromptScorer

scorer = PromptScorer()
prompt = "What is machine learning?"

analysis = scorer.analyze(prompt)
# Returns: specificity_score=2/10, clarity_score=5/10, constraints_score=0/10
# Issues: too vague, no context, no output format specified

Step 2 — Apply Recommendations

Before (score: 2/10):

What is machine learning?

After (score: 9/10):

You are an AI tutor explaining machine learning to a business executive.

Define machine learning in exactly 2 paragraphs:
1. Core concept and why it matters
2. Real-world business example

Use simple language; avoid technical jargon.
Output format: plain text, no markdown.

Step 3 — Few-Shot Pattern

prompt_refined = """
Classify customer feedback as Positive, Neutral, or Negative.

Examples:
- "Great product, fast shipping!" → Positive
- "It works okay" → Neutral
- "Terrible quality, won't buy again" → Negative

Feedback: "The delivery was late but the item is good"
Classification: [ONE WORD ONLY]
"""

Step 4 — Measure Impact

before_accuracy = 0.72  # Original prompt
after_accuracy = 0.89   # Refined prompt
token_reduction = 0.18  # 18% fewer tokens

impact = {
    "accuracy_gain": "+23.6%",
    "token_savings": "-18%",
    "latency_improvement": "-12%",
}

Scoring Rubric

WAF Alignment

Compatible Plays

• Play 01 — Enterprise RAG (refine retrieval prompts)
• Play 03 — Deterministic Agent (structured output)
• Play 18 — Prompt Optimization

Overview

This skill provides a structured, repeatable procedure for analyzes and improves prompt quality with specificity scoring and recommendation engine.. It can be used standalone as a LEGO block or auto-wired inside solution plays via the FAI Protocol.

Category: Prompt Engineering Complexity: Medium Estimated Time: 10-30 minutes

Parameters

Steps

Step 1: Validate Prerequisites

Verify all required tools, credentials, and dependencies are available.

# Check required tools
command -v node >/dev/null 2>&1 || { echo 'Node.js required'; exit 1; }
command -v az >/dev/null 2>&1 || { echo 'Azure CLI required'; exit 1; }

Step 2: Load Configuration

Read settings from the FAI manifest and TuneKit config files.

# Load from fai-manifest.json if inside a play
CONFIG_DIR="${config_path:-config}"
if [ -f "fai-manifest.json" ]; then
  echo "FAI Protocol detected — auto-wiring context"
fi

Step 3: Execute Core Logic

Perform the primary operation: analyzes and improves prompt quality with specificity scoring and recommendation engine..

Step 4: Validate Results

Verify the output meets quality thresholds and WAF compliance.

# Validate output
if [ "$?" -eq 0 ]; then
  echo "✅ Skill completed successfully"
else
  echo "❌ Skill failed — check logs"
  exit 1
fi

Output

WAF Alignment

Compatible Solution Plays

• Play 18

Error Handling

Usage

Standalone

# Run this skill directly
npx frootai skill run fai-boost-prompt

Inside a Solution Play

When referenced in fai-manifest.json, this skill auto-wires with the play's context:

{
  "primitives": {
    "skills": ["skills/fai-boost-prompt/"]
  }
}

Via Agent Invocation

Agents can invoke this skill using the /skill command in Copilot Chat.

Configuration Reference

{
  "skill": "skill-name",
  "version": "1.0.0",
  "timeout_seconds": 300,
  "retry_attempts": 3,
  "log_level": "info"
}

Monitoring

Track skill execution metrics:

Troubleshooting

Notes

• This skill follows the FAI SKILL.md specification
• All outputs are deterministic when dry_run=true
• Integrates with FAI Engine for automated pipeline execution
• Part of the Prompt Engineering category in the FAI primitives catalog