AI Prompt Researcher

Discover high-value AI search prompts your target audience uses on ChatGPT, Perplexity, Gemini, and Claude. Research and generate comprehensive prompt lists...

Security Scan

The skill's files and instructions are coherent with its stated purpose (generating and clustering AI-search prompts) and it does not request credentials or perform networked actions in the included code, but the runtime docs reference additional scripts that are not present.

Purpose & Capability

The name/description (AI prompt research) aligns with the included assets: SKILL.md, a prompt-generation Python script, and extensive reference docs and templates. The Python code produces prompt lists, clusters, and a markdown report — exactly what the skill claims to do.

Instruction Scope

SKILL.md gives concrete instructions and expects the agent to collect brand/category/audience input and run provided scripts. However, the documentation and examples reference additional scripts (scripts/competitor_prompts.py, scripts/trending_prompts.py, scripts/monitor_prompts.py) that are not present in the file manifest. That mismatch is an inconsistency: the instructions advertise functionality that is not shipped. Otherwise, the instructions do not ask the agent to read unrelated system files or access secrets.

Install Mechanism

No install spec is declared (instruction-only + one included script). Nothing is downloaded or written by an install step. The included Python script uses only standard library imports (argparse, json, sys, datetime) and does not pull remote code at install time.

Credentials

The skill declares no required environment variables, no credentials, and no config paths. The provided code likewise does not reference any environment variables or secret material. The level of access requested is proportional to the stated purpose.

Persistence & Privilege

Flags show always:false and default autonomous invocation allowed (normal). The skill does not request permanent system presence or attempt to modify other skills or system-wide settings in the shipped files.

Assessment

This skill appears to do what it says: generate and cluster prompt lists and output a markdown report. Before installing, note that SKILL.md mentions extra scripts (competitor_prompts.py, trending_prompts.py, monitor_prompts.py) that are not included — ask the publisher whether those tools are intentionally omitted or will be added. Also consider: the shipped Python script is local and does not call external services or require credentials, but if missing monitoring scripts are later added they might contact external endpoints — review any added scripts for network calls or API keys. If you plan to run the script, run it in a controlled environment and inspect any new files before giving it broader privileges.

Methodology by GEOly AI (geoly.ai) — in AI search, prompts are the new keywords.

Research and generate AI search prompts that your target audience uses when seeking products, services, or information in your category.

Quick Start

Generate a prompt research report:

python scripts/research_prompts.py --category "<category>" --brand "<brand>" --output report.md

Example:

python scripts/research_prompts.py --category "project management software" \
  --brand "Asana" \
  --competitors "Monday.com,Notion,Trello" \
  --output asana-prompts.md

Why Prompt Research Matters

In traditional SEO, we optimize for keywords. In AI search (GEO), we optimize for prompts.

Key differences:

• Keywords: Short, fragmented (best crm)
• Prompts: Natural language questions (what's the best CRM for a 10-person sales team?)

Understanding the prompts your audience uses helps you:

• Create content that answers those specific questions
• Monitor brand visibility across AI platforms
• Identify content gaps vs. competitors

The 5 Prompt Types

Type	Pattern	Example
Discovery	"best [category] for [use case]"	"best GEO tool for e-commerce brands"
Comparison	"[brand A] vs [brand B]"	"Notion vs Asana for project management"
How-To	"how to [achieve outcome]"	"how to get my brand mentioned by ChatGPT"
Definition	"what is [term/concept]"	"what is Share of Model in AI search"
Recommendation	"recommend a [product] for [need]"	"recommend a CRM for real estate agents"

Full taxonomy: See references/prompt-taxonomy.md

Research Methodology

Step 1: Gather Context

Collect from user:

• Brand name: Your company/product
• Category: Industry/product category
• Target audience: Who buys your product
• Core use cases: Primary jobs-to-be-done
• Competitors: 3-5 main alternatives
• Key features: Differentiating capabilities

Step 2: Generate Prompts

Create prompts across 4 awareness stages:

Stage	User Mindset	Example Prompts
Problem-aware	"I have a problem"	"how to manage remote teams", "why are projects always late"
Solution-aware	"I need a solution"	"best project management software", "tools for team collaboration"
Product-aware	"I'm considering options"	"Asana vs Monday.com", "Notion for project management"
Brand-aware	"I know about you"	"Asana pricing", "does Asana have time tracking"

Step 3: Score & Prioritize

Each prompt gets scored on:

Dimension	Scale	Factors
Intent	Info → Commercial	Likelihood to convert
Volume	Low → High	Estimated query frequency
Competition	Low → High	Difficulty to rank
Value	Low → High	Business impact if won

Priority tiers:

• 🔴 High: Commercial intent + high value
• 🟡 Medium: Mixed intent + moderate value
• 🔵 Low: Informational + awareness building

Step 4: Cluster by Theme

Group related prompts into clusters:

Pricing Cluster
├── "asana pricing"
├── "asana vs monday.com cost"
├── "is asana free"
└── "asana enterprise pricing"

Integration Cluster
├── "asana slack integration"
├── "asana google calendar sync"
└── "asana api documentation"

Output Format

Research Report Structure

# AI Prompt Research Report

**Brand**: [Name]  
**Category**: [Industry]  
**Date**: [YYYY-MM-DD]

## Executive Summary

- Total prompts researched: [N]
- High priority: [N]
- Medium priority: [N]
- Low priority: [N]
- Topic clusters: [N]

## 🔴 High Priority Prompts

| # | Prompt | Type | Intent | Best Platform |
|---|--------|------|--------|---------------|
| 1 | "best [category] for [use case]" | Discovery | Commercial | ChatGPT, Perplexity |
| 2 | "[brand] vs [competitor]" | Comparison | Commercial | Perplexity, Gemini |

## 🟡 Medium Priority Prompts

[Table of informational/commercial mixed prompts]

## 🔵 Low Priority Prompts

[List of awareness-stage prompts]

## Topic Clusters

### Cluster: [Theme]
- [Prompt 1]
- [Prompt 2]
- ...

### Cluster: [Theme]
...

## Platform-Specific Insights

### ChatGPT
- Prompt types that perform well: [list]
- Content format preferences: [description]

### Perplexity
- Prompt types that perform well: [list]
- Citation behavior: [description]

### Gemini
- Prompt types that perform well: [list]
- Unique characteristics: [description]

## Recommended Actions

1. [Action item 1]
2. [Action item 2]
3. [Action item 3]

## Monitoring Setup

Add these prompts to your GEO monitoring dashboard:
- [Tool recommendation]
- [Tracking methodology]

Advanced Usage

Competitor Prompt Analysis

Research what prompts mention competitors but not you:

python scripts/competitor_prompts.py --brand "YourBrand" \
  --competitors "CompetitorA,CompetitorB" \
  --category "your category"

Trending Prompts

Identify emerging prompt patterns:

python scripts/trending_prompts.py --category "your category" --days 30

Prompt Monitoring

Set up ongoing monitoring:

python scripts/monitor_prompts.py --prompts-file prompts.json --frequency weekly

Tools & Resources

• Google "People also ask": Real user questions
• AnswerThePublic: Query visualization
• AlsoAsked: PAA expansion
• Perplexity: Test how prompts are answered
• ChatGPT: Explore prompt variations

See Also

• Prompt taxonomy: references/prompt-taxonomy.md
• Platform differences: references/platform-guide.md
• Prompt templates: references/prompt-templates.md
• Research examples: references/examples.md