Touch people. Turn data into gold.
AI agent skills that extract wealth signals from human behavior, conversations, and relationships.
In 1848, they rushed to California with pickaxes. In 2026, they install skills.
A Golden Touch Skill does two things:
1. It touches people. It reads human signals — conversations, behaviors, decisions, patterns, relationships. The input is always about people: what they say, what they do, what they buy, who they know, how they think, how they make money.
2. It turns data into gold. The output is always a wealth signal — a money opportunity, a deal structure, a market gap, an investment pattern, a revenue path. Not a chart. Not a dashboard. Not a report. A signal you can act on.
King Midas touched things and they turned to gold. Golden Touch Skills touch human data and extract wealth signals from it.
Skills that reverse-engineer how people make money.
| Skill | What It Does | Platforms |
|---|---|---|
| Midas Skill | Feed it any person — it extracts their wealth operating system across 6 dimensions: money engine, deal architecture, investment thesis, risk profile, leverage model, exit pattern. Feed it your own daily data (chats, photos, meetings) — it finds money signals in the noise. Self-learning: gets sharper every input. | OpenClaw · Hermes · Claude Code |
ClawHub · Named after King Midas — 2,700 years of "touch turns to gold," now as an AI skill.
Skills that distill how specific people think and operate.
| Skill | What It Does | Platforms |
|---|---|---|
| President Skill | US presidents distilled into thinking frameworks — mental models, decision patterns, expression DNA. Each president is a perspective you can activate to approach any problem. | OpenClaw · Claude Code |
Skills that find deal opportunities from human interactions.
| Skill | What It Does | Platforms |
|---|---|---|
| Submissions welcome | Skills that scan client conversations, sales calls, or partnership discussions and flag deal signals — who's ready to buy, what pain point keeps coming up, where the leverage is. | — |
Skills that find unmet needs from what people say and do.
| Skill | What It Does | Platforms |
|---|---|---|
| Submissions welcome | Skills that analyze complaints, reviews, support tickets, or social media chatter and surface market gaps — problems nobody is solving yet, services that are overpriced, needs that are underserved. | — |
Skills that extract value from relationship patterns.
| Skill | What It Does | Platforms |
|---|---|---|
| Submissions welcome | Skills that analyze your contacts, interactions, and social graph to find hidden leverage — who connects you to opportunity, which relationships are underutilized, where your network has gaps. | — |
Skills that turn spending behavior into investment signals.
| Skill | What It Does | Platforms |
|---|---|---|
| Submissions welcome | Skills that analyze purchasing patterns (yours or a market's) and extract signals — what categories are growing, where money is flowing, what people are willing to pay more for. | — |
Skills that decode how competitors and industry players make money.
| Skill | What It Does | Platforms |
|---|---|---|
| Submissions welcome | Skills that take public information about a company or competitor and reverse-engineer their revenue model, pricing strategy, cost structure, and growth mechanics. | — |
✅ People input — The skill works with human-generated data:
- Conversations (chat logs, meeting notes, call transcripts)
- Behaviors (spending patterns, browsing history, app usage)
- Decisions (investment choices, career moves, deal terms)
- Content (social posts, reviews, complaints, praise)
- Relationships (who knows who, who influences who)
- Photos and media (what people capture, where they go)
✅ Wealth output — The skill produces actionable money signals:
- Revenue opportunities ("this pain point has no solution — build it")
- Deal structures ("here's how this person structures every deal")
- Investment patterns ("this sector is getting all the money")
- Market gaps ("everyone complains about this but nobody fixes it")
- Leverage points ("you have access to X, which is worth Y to Z")
- Timing signals ("this opportunity has a 6-month window")
❌ Pure data analysis (charts, graphs, dashboards without action signals) ❌ Generic AI tools (chatbots, writing assistants, code generators) ❌ Financial data feeds (stock prices, market data, news aggregation) ❌ Skills that don't involve human behavior as input
In 700 BC, King Midas ruled Phrygia. The legend says everything he touched turned to gold. For 2,700 years, "the golden touch" has meant one thing in every language: the ability to turn anything into wealth.
In 2026, AI agent skills can do what Midas did — touch human data and extract gold from it.
The difference: Midas couldn't control his power. His food turned to gold. His daughter turned to gold. It was a curse.
Your skills are controllable. You choose what to feed them. You choose which signals to act on. The curse became a skill.
In 1848, gold was in the ground. In 2026, gold is in the data.
The people who got richest in the 1848 Gold Rush weren't the miners. They were the ones who sold pickaxes. This list is the pickaxe catalog.
Golden Touch Skills work across agent platforms:
| Platform | Best For |
|---|---|
| Hermes Agent | Self-learning skills that get sharper over time. Built-in learning loop amplifies every golden touch skill. |
| OpenClaw | Broadest ecosystem. 13,000+ skills on ClawHub. Connects to 50+ messaging platforms. |
| Claude Code | Terminal-first. Great for developers who want to analyze data locally. |
| Codex | OpenAI's agent. Works with skills in standard SKILL.md format. |
| Cursor | IDE-integrated. Good for analyzing code-adjacent business signals. |
Found a skill that touches people and turns data into gold? Submit a PR.
See CONTRIBUTING.md for full guidelines.
Quick version:
- Fork this repo
- Add your skill to the right category in README.md
- Format:
| [Name](github-link) | What it does — be specific about the people input and wealth output. | Platforms | - PR title:
Add [Skill Name] - In your PR: explain what human data it touches and what wealth signal it produces
Two rules:
- Must have people as input
- Must have money as output
- Everything else is flexible
⭐ Star this repo if you believe the richest data is human data.
In 1848, they rushed to California with pickaxes. In 2026, they install skills.
Maintained by @realteamprinz · Creator of Midas Skill · PRINZCLAW