Codex Pain Signal Miner Skill

A Codex skill that finds and ranks startup problems from recent public complaints, requests, workarounds, switching behavior, and spending signals.

Give it a market, audience, workflow, competitor, or startup idea. It searches current public sources, clusters independent evidence, separates pain from actual buying signals, proposes narrow MVP wedges, and creates a polished HTML opportunity report.

What It Does

• Researches current public pain and demand signals
• Opens original sources instead of relying on search snippets
• Deduplicates reposts and repeated incidents
• Clusters signals by actor, job, trigger, and failure
• Scores frequency, urgency, spend, workaround cost, reachability, and evidence confidence
• Shows contradictory and weak evidence
• Proposes a narrow product wedge and business-model hypothesis
• Defines the fatal assumption and a 48-hour validation test
• Creates a responsive standalone HTML report

Installation

npx --yes codex-pain-signal-miner-skill@latest

This installs the skill into:

~/.codex/skills/pain-signal-miner

Restart Codex after installation.

Usage

Find problems in a market:

Use $pain-signal-miner to find evidence-backed startup problems for independent gyms and create the final HTML report.

Validate an idea:

Use $pain-signal-miner in idea-validation mode to test whether small agencies have a painful recurring problem managing client approvals. Search for confirming and disconfirming public evidence.

Mine competitor gaps:

Use $pain-signal-miner in competitor-gaps mode to analyze public complaints, feature requests, switching behavior, and workarounds around [competitor].

Output

The report includes:

1. Market verdict
2. Highest-scoring problem
3. Opportunity leaderboard
4. Original evidence links
5. Workarounds and spending signals
6. Opportunity score breakdown
7. Narrow MVP wedge
8. Fatal assumption
9. 48-hour validation plan
10. Unknowns and contradictions

The report ranks research hypotheses. It does not claim market size, revenue probability, or product-market fit without evidence.

Modes

• quick: up to three focused clusters
• standard: up to five mixed-source clusters
• deep: up to eight clusters with contradiction analysis
• idea-validation: confirm and disconfirm one startup idea
• competitor-gaps: complaints, requests, workarounds, and switching signals
• niche-discovery: underserved jobs within an audience or industry

Manual Installation

git clone https://github.com/Kappaemme-git/codex-pain-signal-miner-skill.git
mkdir -p ~/.codex/skills
cp -R codex-pain-signal-miner-skill/pain-signal-miner ~/.codex/skills/pain-signal-miner

Restart Codex after installation.

License

MIT